WO2020104283A1

WO2020104283A1 - Forming cells in a communications network

Info

Publication number: WO2020104283A1
Application number: PCT/EP2019/081283
Authority: WO
Inventors: Johan-Paul Marie Gerard LINNARTZ; Andries Van Wageningen; Armand Michel Marie Lelkens
Original assignee: Signify Holding BV
Current assignee: Signify Holding BV
Priority date: 2018-11-23
Filing date: 2019-11-14
Publication date: 2020-05-28
Anticipated expiration: 2021-05-23

Abstract

A communications network configured to form a plurality of cells, each cell having a polygonal coverage area in a horizontal plane, the cell arranged to provide a channel for transmitting data to one or more client devices within the coverage area. The system comprising for each cell: a plurality of emitters positioned at a respective corner of the cell, to emit a respective beam of radiation within a footprint defining edges of the cell meeting at the respective corner to provide a sub-area of the cell that together the area of the cell; and a modem connected via a respective branch to each of the emitters to modulate at least part of the data into the radiation of each of the respective beams wherein the modem applies a hard handover when a client device leaves a cell and uses a soft handover when a client device moves inside the cell.

Description

FORMING CELLS IN A COMMUNICATIONS NETWORK

TECHNICAL FIELD

The present disclosure relates to forming wireless cells using multiple beam forming emitters. BACKGROUND

Digital communications networks such as cellular networks are typically formed from a number of base stations, modems, and transceivers. The base stations each sit in the centre of an area called a cell. The cell is the area within which the transmissions from the base station may be picked up by a receiver. When positioned next to each other the cells typically fit together to cover larger areas. Each base station is connected to a respective modem. The modem processes outgoing data signals into waveforms suitable for

transmission via wireless channels, and vice versa, it processes incoming wirelessly received waveforms into data.

When the receiver (such as a client device, mobile phone, etc.), moves from the current cell to the neighbouring cell, then any active communication must be handed over to the modem and base station of that neighbouring cell. Such a handover is typically performed when the client device determines that signals received from the neighbouring cell are stronger than those of the current cell. Alternatively, the base station may detect that a handover is needed based on incoming signals from the client device in the reverse uplink. The handover thus occurs when the power of the transmissions from the current cell drops to its lowest levels, typically furthest from the base station in the centre. More specifically a hard handover is one in which the channel in the current cell is released and only then is the channel in the neighbouring cell engaged. A commonly used approach for handovers is the so-called break before make handover, where the connection to the current or source cell is broken before (or 'as') the connection to the next neighbouring cell is made. Other strategies also exist. Hard handovers are intended to be instantaneous in order to minimize the disruption to the call.

Multi input multi output, or MIMO, is a method which uses multiple transmit and receive antennas to multiply the capacity of a radio link by exploiting multipath propagation. MIMO is an essential element of several wireless communication standards including Wi-Fi, 3G, 4G, and Long Term Evolution (LTE) 4G.

In the particular field of the present application, MIMO specifically refers to a practical technique for sending and receiving more than one data signal simultaneously over the same radio channel by exploiting multipath propagation. In typical radio channel configurations the multiple paths may be introduced using reflections within the environment of the signal. In MIMO systems, a transmitter sends multiple streams by multiple transmit antennas. The transmit streams propagate through a matrix channel which consists of all possible paths between the transmit antennas at the transmitter and receive antennas at the receiver. Then the receiver receives the received signal vectors by the multiple receive antennas and decodes the received signal vectors (typically by inverting the channel matrix) into the separate original information streams. Therefore a plurality of independent channels are created which do not interfere with each other.

As compared to a single-transmitter single-receiver system, using the same amount of signal power a MIMO system can provide higher data rates with fewer

transmission errors and better reliability. This form of MIMO spatial multiplexing is fundamentally different from employing a single, common signal waveform shared among multiple antenna which enhances the performance of a single data signal by beamforming and diversity using appropriate phase shifting, amplitude scaling or selection.

MIMO as a technique can also be implemented using light as opposed to radio. In practice MIMO lends itself more to short range light networks than to typical radio networks: in fact MIMO is most effective if one maintains synchronised signals when sent from different locations. This is harder to implement in reality for a radio network, in which cells cover large distances, than for light networks which typically use small cells and thus can share synchronous signals over shorter distances. An example optical MIMO

communication system uses multiple LED-based transmitters and multiple receivers to transfer parallel streams of data.

Previously in literature, academics have considered three different types of optical (light based) MIMO systems:

The first is l-MIMO: This system is implemented using a single luminaire composed of LEDs which each emit differently colored light. Each LED therefore acts as a different transmitter (having a slightly different frequency). The parallel data streams are thus transmitted over different colors of light. The receiver for this system implements optical filters to recover the signals transmitted over each color. The second is s-MIMO: This system is implemented using multiple luminaires that are placed at different locations (usually on a ceiling). Each luminaire is composed of the same type of LED and thus emits the same color of light. In this case, the data streams are separated spatially, each originating from a different location. A‘camera-like’ receiver, that is a receiver that has the ability to distinguish between light coming from different angles, can then separate the different signal streams and recover data.

The third is h-MIMO: This system is a hybrid of the first two systems. It uses multiple luminaires, each composed of different colored LEDs to transmit signals that are separated in both color and space. A‘camera-like’ receiver that can distinguish different colors can then separate the different signal streams and recover the data.

When considering MIMO signals, it is possible for a client device to transition between the multiple signals transmitted. This type of transition is often referred to as a soft handover. For example, this type of handover is performed in typical cellular networks when a receiver (e.g. a client device or mobile phone) moves circularly around a single base station. This is because MIMO in such a configuration is used to supply multiple signals in the form of multiple beams which emanate from the same base station and the same modem. Each signal is emitted to cover a slightly different section of the base station centric cell, but also with some overlap. This configuration may also be referred to as a sectorized cellular network, where the base station has multiple directional antennas which each cover only a separate part or sector of the cell.

In a type of soft handover the current channel is retained and the signals from both the current sector and the neighbouring sector are used in parallel with each other within that channel. That is to say that the connection to the neighbouring sector and corresponding transmitted signal is established before the connection to the current sector is broken. In fact both connections may be maintained for some time, depending on the extent of overlap of their respective coverage areas and the location of the client device. This type of handover may be referred to as a make before break handover. When a communication (e.g. an exchange of data) is in a state of soft handover, the signal which is the best of all those received can be the signal used at a given moment, or some or all of the signals can be combined to produce a clearer copy of the signal. The latter is particularly advantageous, and is easily facilitated by the MIMO technique described above. Soft handovers may also be used between different cells. What makes the handover soft is the absence of a sudden transition between receiving data through one signal to receiving data through a different signal. Modulation methods in digital communications refer to different ways in which digital data may be encoded into the transmitted signal. Examples of these include single carrier modulation, such as Pulse Amplitude Modulation (PAM), orthogonal frequency-division multiplexing (ODFM), and code-division multiple access (CDMA). OFDM for example, is a method of encoding digital data on multiple carrier frequencies. Numerous closely spaced orthogonal sub-carrier signals with overlapping spectra are emitted to carry data. Each sub-carrier signal is modulated with a conventional modulation scheme (such as quadrature amplitude modulation or phase-shift keying) at a low symbol rate.

Because the sub-carriers are orthogonal to each other they do not interfere with each other.

These modulation methods can be used in combination with time-division multiple access (TDMA), in which different signals each only use a fraction of the time.

Any of these modulation and multiple access methods can be used in combination with MIMO, in particular spatial multiplexing MIMO. Yet with different propagation times of the signals from different positions, OFDM is often preferred.

A prior art cellular communication network is disclosed in US patent

6,052,599 which discloses how capacity in a cell is increased by placement of antennas transmitting different signals on the same carrier frequency at spatially separate locations in the cell, other than the center of the cell. Also disclosed is a receiver that uses multiple antennas and uses spatial diversity to extract the desired signal.

The above example MIMO systems can all be used to implement FiFi systems. FiFi, or light fidelity, is a technology for transmitting data at high speed between devices using light. Simply put, Fi-Fi is a visible light communications system that is capable of transmitting data at high speeds over the visible light

spectrum, ultraviolet and infrared radiation. Although the ultraviolet and infrared radiation is not visible to the human eye, the technology for utilising these regions of the spectra is the same. In many instances there are advantages to using these frequency ranges because they are not visible to the human eye.

SUMMARY

Existing technology in this field may experience one or more of a number of issues. For example, for hard handovers between base stations (as per the above explained typical configuration) there is a relatively large range in which the signal to interference ratio requires protection. Only one base station can use the spectrum resources, the other must stay silent. This renders the cellular network inefficient with regards to reuse. Similarly, between base stations the weakness of the corresponding signals emitted by the base stations can lead to poor performance. Diversity and MIMO techniques tout performance gains if used.

However, these gains are often limited in the case where the signals arrive from nearly the same location (for example in optical communications). Thus the gains when a receiver travels around a base station of the configuration described above (where the base station is in the centre and the different MIMO signals are emitted in different directions outward from the base station) are not as good as they could be.

The present application describes a system and method for addressing such issues. The system comprises a cell type structure for use in and as a communications network to provide data to a client device. Each cell has a polygonal shaped coverage area and is formed by multiple emitters which are positioned at different comers of the cell. The beams emitted from the emitters are directed inwards towards the centre of the cell with an opening angle on either side of the comer corresponding to the edges of the cell. Each beam, when positioned at the comers of the cell, has its own sub-coverage area which covers a part or sector of the cell coverage area, but not all of it. When the sub-coverage areas of the emitters of the cell are combined they cover the entire coverage area of the cell.

Hence, according to a first aspect disclosed herein, there is provided a communications network comprising a plurality of cells, each cell having a polygonal coverage area in a top-down 2D projection in a horizontal plane, the cell arranged to provide a channel for transmitting data to one or more client devices within the coverage area; the system comprising, for each cell: a plurality of emitters, each emitter positioned at a different respective comer of the polygonal cell, and configured to emit a respective beam of radiation directed inwards with a footprint in said plane defining edges of the cell meeting at the respective comer to provide a respective coverage sub-area so that the sub-areas combine to extend over the coverage area of the cell; and a modem associated with a separate cell wherein the modem is connected via a respective branch to each of the emitters of the associated cell, and being thereby operable to modulate at least part of the data into the radiation of each of the respective beams of the associated cell.

The plurality of transmitters forming the cell are each positioned at a different physical location to the other transmitters of the same cell, all facing inwards. An advantage of the disclosed arrangement is that the cell has a more sharply defined cell boundary.

In accordance with the invention a first modem of a first cell of the communications network is configured to: perform said modulation using a modulation method for simultaneous multibeam reception, and to send different parts of the data simultaneously over the channel, via said respective beams of the multiple emitters from different respective directions, such that the simultaneous beams are received at the client device along multiple respective paths from different respective angles.

The first modem further is configured to perform a hard handover for a client device associated with the first modem to a second modem, when the client device moves from a first cell associated with the first modem to a second cell associated with the second modem and a soft handover of the client device associated with the first modem when within the first cell, by adapting a proportion of the data transmitted to the client device via respective sub-channels of respective emitters of the same cell, such that the proportions of the data emitted are varied.

In embodiments the plurality of transmitters may have an equal number of spatially co-located receivers having a directional sensitivity or aperture substantially corresponding with that of the emission profile. In this manner the communications network (100) can enable bi-directional communication.

In embodiments the footprint of the beam from each emitter has a sectorized shape, i.e. two sides emanating from the emitter with an equilateral opening angle (the shape of the sector of a circle, except not necessarily limiting to a particular shape at the end of the beam).

In embodiments, the modem is configured to perform said modulation using orthogonal frequency division multiplexing. Preferably, the modulation method is orthogonal frequency-division multiplexing, OFDM, or code-division multiple access, CDMA.

The advantage of these types of modulation methods it that they are particularly suited to allowing the simultaneous reception of multiple signals in the same bandwidth or wavelengths range, and preferably tolerant to small timing offsets.

In embodiments, the polygonal coverage area comprises at least four comers.

In embodiments, the polygonal coverage area comprises a square.

In embodiments, the polygonal coverage area comprises at least six comers.

In embodiments, the polygonal coverage area comprises a hexagon.

In embodiments, the radiation comprises: light or line-of-sight high frequency radio.

In embodiments, the radiation is radio, and each emitter is configured to emit the respective beam by beamforming using an array of emitting elements.

In embodiments, the radiation is light, and each emitter is configured to emit the respective beam of light by beamforming using an array of emitting elements. The advantage of using light is that light from the emitters can be accurately collimated using optical elements such that the coverage area of a cell can be covered in an accurate manner, by aligning the respective coverage sub-areas of the transmitters with the boundaries of the polygonal cell area.

In embodiments, the radiation is light, and each emitter is configured to emit the respective beam of light with distribution controlled by one or more optical components.

In embodiments, the client device is transferred from an emitter of the first cell located on a base station to an emitter of the second cell located on the same base station.

In embodiments, the hard-handover is performed only at the cell boundary.

In embodiments, the cell boundary is between beams of two emitters located on the same base station.

In embodiments, the network comprises the client device.

Preferably the abovementioned client device comprises a multi-directional receiver. In contrast to a prior art MIMO directional receiver where such directional receiver is used to spatially distinguish between different access points (each with an individual modem) transmitting data to the receiver, the multi-directional receiver as presented herein is arranged to receive data that originates from multiple directions simultaneously, but which originates from a single modem. As a result the signal contributions from ah receivers may be forwarded to the client-device MIMO demodulator for further processing.

The multi-directional receiver according to the invention may be placed in a dongle for providing connectivity to already existing equipment, or may be integrated in devices, such as laptops, monitors, or mobile devices.

Preferably such a multi-directional receiver is arranged to receive light or radio frequency signals from more than two spatial directions. More preferably the multi directional receiver is tailored to the maximum number of transmitters it may receive a signal from at any position within a cell. For example, consider the situation wherein a receiver is placed on a table at the centre of a cell having «-transmitters; at this point in the cell the multi-directional receiver may receive signal contributions from each of the n ceiling-based transmitters.

Preferably the number of directions that the multi-directional receiver can receive signals from is equal to the typical number of transmitters in the polygonal cell and the opening angle for receiving a signal is tailored such that contributions from a single transmitter may be received (so in case of «-transmitters the opening angle in the horizontal plane would be 360 In).

When oriented correctly, the signal received may substantially correspond to a single transmitter/transceiver. When the sensor apertures or antennae are not properly aligned, contributions from multiple transmitters/transceivers may be received. In case of light, the number of transmitters would but would still be limited to two transmitters (worst cast - assuming no sensor tilt). When there are multiple signal contributions present in the sensed signals, these contributions can in turn be handled by the MIMO modem at the receiving device.

By decreasing the number of directions beyond the number of transmitters, the number of transmitters/transceivers contributing to the received signal of each of the directional sensors will increase. By further increasing the number of directional sensors the spatial selectivity may further improve. Increasing the number of receivers may be particularly beneficial when device tilt is also to be taken into account as is envisaged for mobile devices exhibiting (erratic) motion and/or rotation along various axis.

In case bi-directional communication is required and the ceiling-based transmitter devices are in fact transceiver devices, then the aforementioned multi-directional receiver of the client-device would conversely also need to be equipped with a multi directional transmitter.

According to a second aspect disclosed herein, there is provided a method of forming a communication network comprising a plurality of cells, each cell having a polygonal coverage area in a top-down 2D projection in a horizontal plane, the cell arranged to provide a channel for transmitting data to one or more client devices within the coverage area; the method comprising, for forming each cell: positioning each emitter at a different respective comer of the polygonal cell to emit a respective beam of radiation directed inwards with a footprint in said plane defining edges of the cell meeting at the respective comer to provide a respective coverage sub-area so that the sub-areas combine to extend over the coverage area of the cell; and connecting a modem associated with a cell to each of the emitters of the associated cell via a respective branch of the modem, and being thereby operable to modulate at least part of the data into the radiation of each of the respective beams.the method comprising a first modem of a first cell of the communications network: performing said modulation using a modulation method for simultaneous multibeam reception, and to send different parts of the data simultaneously over the channel, via said respective beams of the multiple emitters from different respective directions, such that the simultaneous beams are received at the client device along multiple respective paths from different respective angles.

The method further comprises the first modem performing a hard handover for a client device associated with the first modem to a second modem, when the client device moves from a first cell associated with the first modem to a second cell associated with the second modem and performing a soft handover of the client device associated with the first modem when within the first cell, by adapting a proportion of the data transmitted to the client device via respective sub-channels of respective emitters of the same cell, such that the proportions of the data emitted are varied.

In embodiments the network comprises a filler cell, the filler cell having its own modem and being located between some or all of the plurality of cells, thereby providing an intermediate connection to bridge any gap between the plurality of cells during a hard- handover.

According to a third aspect disclosed herein, there is provided a modem for use in a communications network comprising a plurality of cells, each cell having a polygonal coverage area in a top-down 2D projection in a horizontal plane, the cell arranged to provide a channel for transmitting data to one or more client devices within the coverage area; the system comprising, for each cell: a plurality of emitters, each emitter positioned at a different respective comer of the polygonal cell, and configured to emit a respective beam of radiation directed inwards with a footprint in said plane defining edges of the cell meeting at the respective comer to provide a respective coverage sub-area so that the sub-areas combine to extend over the coverage area of the cell, the modem for association with a separate cell wherein the modem is connected via a respective branch to each of the emitters of the associated cell, and being thereby operable to modulate at least part of the data into the radiation of each of the respective beams of the associated cell.

The modem when associated with a first cell is configured to: perform said modulation using a modulation method for simultaneous multibeam reception, and to send different parts of the data simultaneously over the channel, via said respective beams of the multiple emitters from different respective directions, such that the simultaneous beams are received at the client device along multiple respective paths from different respective angles Moreover the modem is further configured to perform a hard handover for a client device associated with the first modem to a second modem, when the client device moves from a first cell associated with the first modem to a second cell associated with the second modem and perform a soft handover of the client device associated with the first modem when within the first cell, by adapting a proportion of the data transmitted to the client device via respective sub-channels of respective emitters of the same cell, such that the proportions of the data emitted are varied .

According to a fourth aspect disclosed herein, there is provided a kit comprising a base station and a plurality of modems according to the third aspect disclosed herein above for use in a communication network comprising a plurality of cells, each cell having a polygonal coverage area in a top-down 2D projection in a horizontal plane, each cell arranged to provide a channel for transmitting data to one or more client devices within the coverage area; the base station comprising: a plurality of emitters, each emitter configured to emit a respective beam of radiation to provide a respective coverage sub-area of a different respective one of said cells, the sub-area constrained by an opening angle of the emitter such that the sub-areas combine to extend around the circumference of the base station in a horizontal plane, each emitter configured to receive a signal from a different modem for outputting to a different cell; and wherein the base station is configured to enable each of the emitters to be controlled by a different modem of the plurality of modems.

BRIEF DESCRIPTION OF THE DRAWINGS

To assist understanding of the present disclosure and to show how embodiments may be put into effect, reference is made by way of example to the

accompanying drawings in which:

Fig. 1 shows a single network cell comprising three emitters and one modem;

Figs. 2A and 2B show respectively, the plan view and the side view of an emitted beam;

Fig. 3 shows a one dimensional network formed of two cells;

Fig. 4 shows a base station for forming a two dimensional network;

Fig. 5 shows a two dimensional network comprising twelve cells formed by six base stations;

Fig. 6 shows the six base stations of Figure 5 with the connecting cabling from the respective emitters to each respective modem;

Fig. 7 shows a one dimensional network with a filler cell;

Fig. 8 shows a base station for a two dimensional network with a filler cell;

Fig. 9 shows a side-view of a one dimensional cell comprising two transmitters and a client-device comprising a multi-directional receiver; Fig. 10A shows a plan- view of two-dimensional network comprising four transmitters/transceivers in a rectangular cell and one well aligned client-device comprising a multi-directional receiver sensing in 4-quadrants;

Fig. 10B shows a plan- view of a two-dimensional network comprising four transmitters/transceivers in a rectangular cell and one poorly aligned client-device comprising a multi-directional receiver sensing in 4-quadrants;

Fig. IOC shows a plan- view of a two-dimensional network comprising four transmitters/transceivers in a rectangular cell and one client-device comprising a multi directional receiver sensing with two sensors sensing opposite halves of the space around it.

DETAILED DESCRIPTION OF EMBODIMENTS

If users move across the cell coverage boundary of a base station in a cellular communication network, a handover is needed. In the original set up of wireless radio networks, each base station covers a region around it. Such handovers (hard handovers) occur in between the base stations. It is a protocol action to switch from one base station to another.

Increasingly it is seen that modulation methods exist that that allow multicasting or other forms of simultaneous transmission of one or multiple signals from multiple antenna’s on the same channel and at the same time. In this manner these signals can be considered to jointly improve the link quality. These include OFDM Single Frequency Networks SFN in Broadcasting (DVB, DAB), in which the same signal sent from two or more antenna locations can be received sometimes many kilometers apart. Another example is Multiple Input Multiple Output or MIMO, in which parts of the bit stream are sent separately over multiple antenna’s, each part having a different propagation path to the receive antenna’s. This is typically combined with adaptive beamsteering.

In this document forms of such simultaneous transmission as described above are utilized, but sent from spatially separated locations such as different base station locations. In doing this the technique of MIMO is exploited, but the different propagation paths are created by sending the signals from different locations. This avoids the need to rely on reflections in the environment to create the different paths as in typical MIMO radio configurations, and makes it suitable for optical communication.

This document describes a system in which hard handovers do not occur in between base stations, but rather near the base stations. Instead, the transitions in between the base stations are addressed via a soft handover, using signals on the same channel and appropriate modulation methods. In other words, the coverage area of a particular cell in which the user communicates is no longer located around a base station, but is now located in between base stations.

It is recognised herein that when handovers occur halfway between two base stations, there is a relatively large range in which the signal to interference ratio requires protection. Only one base station can use the spectrum resources, the other must stay silent.

It has been recognized that a way to deal with this problem in a cellular system is to design the cell layout in such way that the signal is strong inside the cell, but it drops off very rapidly at the edges of the cell. This rapid drop off can be achieved best near the base stations, or between sectors of the same base station. While in between the base stations, a gradual transition (soft handover) is used where the client device is getting the data signal from multiple base stations, the signals sharing the same frequency and the same time periods.

Further, in the standard configuration, handover between the cells (and hence halfway between the base stations), involves determining at any one time which cell or base station the client device should be connected to. This can present problems as the limits of the signals in these places are defined predominantly by transmit power of the signal. As a result there can be a lot of cross-talk halfway between the base stations.

By implementing the above described configuration of a cell, one where the cell is the area between the base stations rather than the area around a single base station, cross-talk halfway between two base stations becomes beneficial. This is because these signals all belong to the same cell, and the client device benefits from the signals transmitted from the multiple base stations combining to reinforce each other (via soft handover and MIMO). Similarly, with physical layer multicasting or spatial multiplexing, the signal is strong inside the cell, but drops off very rapidly outside the cell. This creates well defined cell boundaries, facilitating hard handovers between cells by creating a lot less cross-talk between the signals of different cells.

Thus, in a cellular system with the ability to do a hard handover and a soft handover (e.g. in the form of multicasting, site diversity or MIMO), one can deviate from the common practice of locating cell boundaries somewhere halfway between the base stations.

Diversity and MIMO as concepts promise performance gains, but in the typical optical network configuration the gains are limited. This is because the signals arrive from nearly the same location, from the same base station in the centre of the cell. The signals experience correlated channel statistics. By implementing the above described configuration of a cell, one where the cell is the area between the base stations, MIMO is used to perform soft handovers in between different base stations which are not at the same location. As a result the gains promised from MIMO can be realized even for optical links, and exploited at the weakest point, i.e. in the centre of the cell where the gradual limit of the signal due to transmit power is located. In the presently described case, the hard handovers are predominantly executed near edges, between signals coming from the same base station, and not in cell centres.

The above described system is particularly suited for optical communication in very small cells, such as LiFi. This is because a link from one single modem to multiple base stations is required to create the cell between the base stations. This link can maintain better synchronization between the emitted signals and therefore the signals as received by the client device within the cell if the signals experience the smallest amount of delay possible. Similar suitability exists for radio signals in the millimeter wavelength range, that is above a few (tens of) GHz.

Known multicasting solutions (e.g. SFN, MIMO, STBC) work best if the signals have the same symbol timing. Thus the synchronization of the signals should not be lost when sending them from the modem to the base stations. Typically a fiber optic cable carrying an analog waveform can ensure this. A digital connection in which the signal is converted back into bits and data packets would run the risk of losing such precise synchronization because of the unknown, unpredictable, and differently varying processing times along each route. In such a case, an accurate distributed clock would be needed. Thus, a digital exchange is possible but not ideal.

Figure 1 shows a communication network 100 configured as described above and formed from at least one such cell 102. The cell is arranged to provide a channel for transmitting data to one or more client devices 108. The client device may be for example, a dongle, smart phone, tablet, internet of things (IoT) device, etc. that is able to connect wirelessly to the channel. The cell 102 comprises three emitters 104a-c. Each emitter 104a-c is positioned at a different comer of the polygonal cell. In this example the cell 102 is hexagonal in shape. However, the cell may be the shape of any polygon. In particular, tessellating polygons may be used (e.g. hexagons, squares, etc.), such that planar surface areas can be covered without overlap of cells or leaving gaps between the cells.

Combinations of multiple types of polygon may be used within the same network to achieve the required coverage. The polygon may have any suitable number of comers. The polygon may have at least three comers, at least four comers, or at least six comers, etc. The polygon(s) in the network may be any of a square, a triangle, a hexagon, a rectangle, a pentagon or any combination thereof.

For instance in lighting, e.g. on ceilings, triangle and hexagon shaped cells can be made using a rectangular grid of a ceiling tiles. For example, one out of two (or four) tiles may contain a LiFi lighting unit, and then alternating a shift of the LiFi units (e.g. for an odd row: position 1, 3, 5; an even row: position 2, 4, 6), gives a network with hexagonal cells.

Each emitter 104a-c emits a beam of radiation 106a-c. Each of the beams 106a-c are directed inwards. That is to say that the beams are directed toward the centre of the polygonal cell. The centre of the cell in this configuration is located centrally between the emitters of the cell. Each beam 106a-c has a sectorized footprint in the plane of the cell and provides coverage for a sub-area 106a-c of the cell. That is to say each beam covers part of the total coverage area of the cell 102. It should be appreciated that the sub-areas 106a-c of each beam emitted within the same cell overlap each other. In other words the sub-areas are not disjointed. The extent to which the sub-areas of the emitters overlap should be substantial enough to allow for the existence of locations within the cell where signals from more than one of the emitters of the same cell can be received by a client device 108. The overlap exists where the beams from the different emitters of the same cell meet. In this configuration the overlap deliberately occurs at the limits of the signal towards the centre of the cell, i.e. where the transmission power is at its weakest. Therefore in these regions where the signal strength from one emitter is low, the overlapping signals can all be received simultaneously and used in combination to reinforce each other. This is an application of the above describe MIMO technique, where data can be received along different paths but in the same channel without interfering with each other.

Each beam is a sector beam, or has a sectorized footprint 106a-c. The edges of the cell 102 are hence defined by the outer edges of the footprints 106a-c, the outer edges being those which meet at the comer from which the beam is emitted, e.g. at the respective emitter 104a-c. The sub-area of each beam may also be referred to as a sector of the cell, hence the use of the term sectorized. The two straight sides defining the opening angle of the emitter are the same shape as the straight sides of a sector of a circle (but not limited to a particular shape of the end of the beam). N.B. opening angle does not necessarily limit to any particular mean of forming the angle, e.g. optics, an aperture, or beamforming by means of an array of antenna elements.

That is to say, the emitters emit a respective sector beam of radiation directed inwards, on frequencies and instances (at least partly) shared with other emitters of the same cell. They jointly cover the entire cell on the shared frequency band, each of the emitters emitting at angles within the span of the two connecting ribs of the polygonal cell.

When the beam of radiation is line-of-sight radio waves, the shape of the footprint or sub-area may be constructed using beamforming. The beamformed radio signal may be constructed using an array of emitting elements, by controlling the relative amplitudes and/or phases of the signals from the different emitting elements. When the beam of radiation is light, the shape may also be constructed using collimation, lenses, or other forms of beamforming. The beamformed light being constructed using an array of light emitting elements (again by controlling the relative amplitudes and phases). Alternatively or additionally, the shape of the light beam’s distribution may be controlled by optical components.

Line-of-sight high frequency radio waves behave in a similar way to light, in that they do not reflect easily like low frequency radio waves and are blocked by obstacles between a transmitted and receiver. That is the line-of-sight is highly dominant over the multipath reception. Thus, if all signals originate from the same base station and all signals arrive at the same client location, the channel matrix, describing the signal strengths and phases between all emitters and all receivers, becomes a very regular structure with almost identical values. Hence such a channel matrix would not have enough strong Eigen values, which is a requirement for effective MIMO reception.

Each emitter of the cell 102 is connected to the same modem 112 via a respective one of multiple branches of the modem 1 lOa-c. The modem is a unit that processes incoming data signals into waveforms suitable for exchange via wireless optical or radio channels. The modem can interface with the wireless channel via multiple branches (input and output). In order to emit its wireless signals, the modem must make use of at least one emitter.

The modem 112 is operable to modulate at least part of the data to be transmitted into the radiation of each of the respective emitter beams. The modem is configured to perform modulation using a modulation method suitable for simultaneous multibeam reception. The modem can use a modulation method that exploits the reception of the signals via multiple paths, such as OFDM, CDMA, and MC (multi-carrier)-CDMA, which facilitates a soft handover between two branches of the same modem.

Thus a client device within the cell may receive the data from any one of the emitters of the cell alone, or any combination of the emitters of the cell simultaneously. The modem is configured to adapt a proportion of the data transmitted to the client device via the respective sub-channels of the respective emitters, and as such the proportions of the data emitted can vary between emitters. When receiving the data from more than one emitter of the cell, the signal received from each emitter may comprise all of the data or only part of the data (the data being all of the data to be delivered to the application of the client device). It should be understood that the coding of the data transmitted by each of the emitters may be different, even if the data itself (the application data) is the same for all of the emitters of the cell. As explained above a soft handover can be thought of as a smooth transition in the flow of data from being via one branch of a modem to being via a second branch of the same modem.

That is to say, the modem is connected via a respective branch to each of the emitters, and is operable to modulate or encode the user data across multiple respective beams, using a modulation method that is tolerant to transmission and reception of multiple beams simultaneously. For example, a variant of OFDM suitable for non-negative transmission over optical wireless channels. That is OFDM with a cyclic prefix sufficiently large to allow signal reception from multiple emitters with different delays.

By arranging the cells of the communications network as shown in Figure 1 , where the emitted beams are normally considered to be weakest, that is at the limits of the individual beams as transmitted from the emitters, the beams are able to be strengthened by using more than one of the beams to cover the same location and simultaneously transmit data. In the middle of the cell the achievable bitrate drops, but MIMO means it is possible to use all transmitters. This is based on the MIMO technique as explained above. Here the technique is to receive parts of the data simultaneously over a channel via respective beams of the multiple emitters. The same data signal is sent, and the modem performs a matrix calculation to split the data (but with the same application data content). Each beam is thus sent from a different respective direction, such that the simultaneous beams are received at the client device along multiple respective paths from different respective angles. The MIMO signals of the same cell are distinguishable from each other in that they are received from different angular directions to the client device. Thus this is based on MIMO but it is not entirely the same as traditional radio MIMO due to a no reflection based aspect.

In practice the emitters of the cell will each be located on a base station (not shown), the base station is a physical object, e.g. a LiFi or RF enabled luminaire in the ceiling, or a OWC or RF system in a street pole, upon which the emitters are mounted.

However, in embodiments, and unlike prior systems, the emitters of a given cell may be located on different base stations (see Figures 3 to 8, discussed in more detail shortly). The receiver 108 need only be a receiver which is sensitive in multiple directions. That is to say the receiver must be able to detect a signal in multiple directions and receive orthogonal signals. This is so that the channel matrix can be inverted. This is most easily achieved by having differently angled receivers. Preferably the receivers are tailored to the typical emitter configuration, for example when using a multi-directional receiver in a square cell, preferably the receiver has four directional receivers, which e.g. when used on a table mounted laptop, can collect signal from four directions. In contrast to prior art MIMO systems where a directional receiver may used to spatially distinguish between different access points (each with a separate modem) transmitting data to an end-user device. The directional receiver as described herein is arranged to receive data that originates from multiple directions simultaneously that all originate from a single modem.

In embodiments the above mentioned modulation method may be single frequency network OFDM which does not split the data, but facilitates diversity reception. The respective beams of the cell may each carry the same waveform via differently delayed respective paths. For example, using Coded-OFDM with a sufficiently long cyclic prefix.

When using high frequency radio waves, the edges of the lobes may not be as tight as can be done for light; i.e. there may be spillage into adjacent areas for radio technology. Light can be collimated into a strict beam, whereas radio phased arrays are less sharp. However, some side lobes (as seen in beamformed radio signals) may be used to help give the MAC protocol time to perform a hard handover.

Figures 2A and 2B show respectively a plan and a profile view of the beam emitted from one emitter. These are only example views, the beam may vary in width, length, and height covered above the x-y plane (along z-axis).

In Figure 2A the edges of the beam parallel to the x and y axis form hard constraints of the shape of the sector. These edges are formed by the beamforming or optical components described above, and result in a sharp drop off in signal strength across the edges of the beam in these regions. The edge of the beam opposite the emitter is determined based on the signal transmit power. As a result the beam here has a gradual drop off in signal strength, and the edge is softly defined. It will be clear to those skilled in the art that the shapes shown in Figures 2A and 2B depict the core lobes of the transmission range, which are more akin to those for radio frequency beamforming; wherein outside of the lobes there will also be some RF signal present, but more attenuated. In contrast beam-forming for light is easier; and a much sharper fall-off can be realized on the edges of the polygon cell. Figure 2B illustrates how the beam may be emitted from an emitter positioned off of the horizontal plane x-y, where edge parallel to the z axis can be thought of as another hard constraint of the beam’s sub-area. The soft edge, furthest from the origin of the x-y-z axes, can depend on the elevation of the transmitted beam and where it intersects with a surface.

Figure 3 shows an example implementation of the network as a 1 dimensional network. For example, such a network may be implemented in a corridor or tunnel, where base stations 302a-c may be positioned in a row. For example, the base stations may be Optical Wireless Communication (OWC) enabled base stations.

The base stations 302a-c each comprise two emitters (e.g. emitters 104b and 104c). The emitters of each base station 302a-b belong to a different cell 102a-b of the network, and are thus connected to different modems 112a-b. The emitters 104b-c function in the same way as those described with reference to Figure 1. However, in this configuration it can be seen that emitter 104c belongs to a different cell 102b to that of emitters 104a-b. The emitters 104a-b, located in different base stations 302a-b, are connected to a single modem 112a. This modem 112a controls the modulation and encoding of all of the data to be transmitted by that cell 102a.

By using the configuration shown in Figure 3, the total coverage area between the emitters belonging to the same cell (the coverage area of the cell) is covered by the sub- areas of each emitter of that cell in combination. When a client device moves from an area covered solely by the beam of the left-hand emitter 104a of the cell 102a, towards the sub- area covered solely by the beam of the right-hand emitter 104b of the cell 102a, it passes through a region covered by both sub-areas. Within this region the client device experiences the call state described above, known as a soft handover. While in this region, the client device may send feedback to the modem 112a of the cell 102a to inform it of which signal is being used and how much. This allows the modem 112a to determine the best proportions of the data to send via the respective signals (e.g. via the respective beams). Soft handover in itself is well known in the art.

When a client device moves from the coverage area of the left-hand cell 102a into the coverage area of the right-hand cell 102b the client device experiences a call event known as a hard handover, as described above. In the hard handover the modem of a first cell transfers the client device from being connected to the first cell to being connected to a different second cell adjacent to the first cell. This hard handover is typically performed at the cell edge. Although communication with multiple modems is used, this is mainly for the handover protocol control and signaling. The hard handover is performed at the MAC layer. The main client application data flow is transferred via one modem at a time.

It can be seen from Figure 3 that this hard handover occurs underneath the base station 302b. This is different to the prior art, where the hard handover, the transition between emitters controlled by the same modem, is executed in between base stations. Using the network and cell configuration described herein, the hard handovers can be seen to be optimally positioned where the signal strength is high and the edges of the beam are well defined. This facilitates the mechanisms of the hard handover by reducing the cross-talk between signals transmitted by emitters of different cells.

In reality there may be some overlap between the coverage of the two different cells near the base station. However this is not entirely bad, since a client device cannot instantly be handed over to a new cell. The configuration as shown in Figure 3 does allow for the hard handover to be executed much faster than typically possibly when executed halfway between base stations where the signal strength is low and the cell edge is harder to determine.

Figure 3 shows separate connections between the modems and each of their respective emitters for the sake of clarity. When implemented, a high speed backbone may connect all of the modems to a central unit. This central unit may then route the data to and from the appropriate modem. When two modems execute a hard handover the data is flow is simply changed to go to the new modem. There are several ways of doing this which are known in the art.

Figure 4 shows a single base station 400 for implementing the desired network configuration in a two dimensional grid. The base station 400 comprises four emitters (1-4). Each emitter emits a beam of radiation providing a coverage sub-area 402a-d. Each of the sub-areas or sectors 402a-d attached to this base station 400 has a coverage sub-area spanning a 45 degree angle radially about the centre of the base station. In combination the separate sectors 402a-d extend to cover an area radially outward and across the whole circumference of the base station in the horizontal plane.

The beam emitted by each emitter (1-4) belongs to a different modem (not shown). Therefore in Figure 4 the sub-area each have a different pattern to represent the different channels used for transmission.

The base station may have more or fewer sectors emitting radially from it. The number of emitters used is determined based on the desired shape of the network cell. For example, referring back to Figure 1, the corresponding base station used to provide a plurality of such hexagonal cells would require three emitters to cover a 360 range around each base station.

Figure 5 shows the base station of Figure 4 implemented in a grid pattern to form a two dimensional network. The network 500 comprises 6 base stations 502a-f. Each base station comprises four emitters (1-4). In this example the two central squares of the grid 6, 7, having a base station on each comer, represent two complete cells of the network 500.

In such a grid pattern the edges or sides of the square cells may be called ribs. Thus in the present configuration all ribs of a complete cell coincide with at least the sectorized emission from two emitters. The boundaries of the sub-areas forming the ribs of the polygonal cell.

The complete square cell on the left 6 has emitter number 4 from base station 502a, number 3 from base station 502b, number 1 from base station 502e, and number 2 from base station 502d transmitting towards its the centre. The hard edges of these respective sub- areas create a well-defined cell boundary to provide for good hard handovers between cells. The soft edges of these respective sub-areas overlap in the centre of the cell 6 and towards the middles of the hard edges. These overlapping sub-areas facilitate a soft handover region where multiple signals can be used simultaneously to compensate for diminished transmit power of the individual beams. The respective emitters of the cell 6 as listed above are each connected to the same modem (not shown), and as such each transmit data on the same channel. This is illustrated in Figure 5 as it is in Figure 4 using a particular pattern to fill the respective sub-areas.

The cell spanning square 7 of the grid is configured similarly to that of the cell spanning square 6. However, as the emitters of this cell 7 are connected to a different modem (not shown) the emitted beam is transmitted on a different channel to that of cell 6. This is illustrated by the different pattern filling the coverage area of cell 7.

Figure 5 also shown a client device 504 with an arrow showing its direction of movement. As the client device crosses from cell 6 into cell 7 any active data session will experience a hard handover event. During this handover the data session will be transferred from the modem of cell 6 to the modem of cell 7.

Should the client device move only within the coverage area of cell 6, the data session will only experience a soft handover state as it moves among the respective beams of the emitters of cell 6. As described above, all of the emitters within a single cell may be active, or specific ones may be active only when needed. The modem of the cell knows which channel is being used and sends data as needed based on feedback from the client device (e.g. a dongle). The method of setting up the communications network described herein, for example as shown in Figure 5, comprises positioning the emitters on two or more comers of the cell. The emitters may be positioned at every comer (as in Figure 5), at alternate comers (as in Figure 1), or some irregular pattern of comers of the desired cell shape such that the desired shape is achieved. The required modem for each cell may then be attached to each emitter of the respective cell via a respective branch of the modem.

The two dimensional configuration shown in Figure 5 (and other such cell shapes), may also work with MIMO enabled modems having more than one channel. For example the data may be split over two channels XI and Y1. In reference to Figure 5 this may be performed by the data sent from emitter number 4 of base station 502a, and emitter number 1 of base station 502e emitting data channel XI, and emitter number 3 of base station 502b, and number 2 of base station 502d emitting data channel Yl. Where the beams from these emitters overlap both channels would be able to be received. This provides for two different data streams for use in one such cell, and in a network comprising four such cells this provides the ability to use eight independent data streams. That means reusing the spectrum eight times. The modem of a cell having two channels would comprise two branches (or sources), each branch having two ports. It may also be possible to have a cell where each emitter emits a different MIMO channel. The respective beam sub-area overlaps would provide whichever channels overlapped at that location to a client device at that location.

Figure 6 shows the respective modems 602a-b of cells 6 and 7 and the respective branches 604a-d, 605a-d of those modems as would be needed to implement the network of Figure 5. Modem 602a uses branch 604a to send a corresponding signal to emitter number 4 of base station 502a, branch 604b to send a corresponding signal to emitter number 3 of base station 502b, branch 604c to send a corresponding signal to emitter number 2 of base station 502d, and branch 604d to send a corresponding signal to emitter number 1 of base station 502e.

If the radiation beams in the network are light, the physical connections between the modem and the emitters are ideally of a similar length. This is because when using a fibre optic cable the length is directly proportional to the time the signal takes to traverse the fibre. In the present configuration it is desirable to maximise the synchronicity of the signals as emitted from the emitters within the same cell. This is to maintain the ability to exploit the multipath configuration introduced within the cell by locating the emitters at the edges of the cell. It can be seen from Figure 6 that wiring (or fiber optics) stretch from one base station to another base station, and all sectors (beams of emitters) on one base station get their signal from a different modem.

The orientation of the modems 602a and 602b as shown in Figure 6 (with the connecting sides of the modem directed towards the base stations and relative emitter to which they are to connect to) is simply to illustrate the connections.

For example, when guiding an installer of the system in correctly connecting the wires between a modem and the corresponding base stations, some example support may be:

Align each connector of a base station with the direction of the coverage area of a sector of the base station. Each connector on a base station may be physically positioned on the base station such that its orientation corresponds to the desired direction of the beam. To facilitate this, or alternatively, each connector and each optical element of the base station may be marked with corresponding symbols, colours, etc.

Align each connector of a modem to the direction of a base station. Each connector of the modem may be physically positioned such that its position indicates in which direction a wire needs to be connected to a base station.

Point b) can be used to avoid cabling errors in the case where two different modulation schemes (e.g. XI and Yl) are used, as described above.

To keep the delay over the cables as equal as possible, the cables between a modem and its base stations should be of equal length. It is thereby assumed that the modem is positioned equidistant between, or near the middle of the space between, the base stations.

Figures 7 and 8 show respectively a one dimensional and two-dimensional implementation of the above described network configuration, with the addition of a further cell. The additional or filler cell is located between the cells of the network in Figure 3 or at and around each of the base stations in the network of Figure 5. The purpose of the filler cell is to facilitate the hard handover between cells. Particularly where either interference or gaps between cells might occur.

In the one dimensional network of Figure 7 it can be seen that there is a gap between the two cells 102a and 102b. This gap is covered only by the filler cell 702. In this one dimensional example the filler cell 702 comprises a single emitter 704. The emitter 704 is connected to its own modem 706.

The filler cell 702 functions such that as a client device approaches the edge of cell 102a it also enters the coverage area of cell 702. At this time the client device may transfer via a hard handover to cell 702. The client device may then cross the gap into cell 102b’s coverage area while connected to cell 702 and thus avoid losing its data connections altogether. If the client device entered the gap and lost its connection to cell 102a, any active data sessions would be dropped. Once the client device determines it has entered the coverage area of cell 102b it may perform another hard handover to transfer to cell 102b.

In the two dimensional network of Figure 8 the beams from the co-located emitters around the same base station 800 are particularly strong near the base station.

However, similarly to the case described above, directly at the base station a gap between the sub-areas of the emitters of the same base station may be deliberately implemented to prevent and safe guard against interference. Thus to support the hard handover at a base station 800, the centre area may be covered with a dedicated filler cell 802, signified in Figure 8 but a dark patterned area 804. The cell may be formed using one or more emitters and a dedicated modem (not shown).

The filler cell 802 functions such that if a client device moves towards a gap at the base station, any active data session can be handed over in a hard handover to the filler cell 802. As soon as the client device detects that it is in the coverage area of another cell, another hard handover can be performed to transfer the client device from the filler cell 802 to that other cell.

Such a central filler cell 802 may also be used to minimize the number of hard handovers performed close to the base station. For example, it may be possible for the client device to travel between different cells of the network in quick succession when close to the base station. By connecting to the filler cell 802, these relatively frequent hard handovers may be avoided by giving preference to staying connected to the filler cell 802 as the client device moves around the base of the base station. Preference to the filler cell may be especially high when a client device is near the centre of the filler cell. This helps to prevent the device losing connection or experiencing so many handovers when moving around the same base station. However, at the edge of the filler cell where its signal is weaker, preference to another cell with a stronger signal should be given.

In general a filler cell may be located between some or all of the plurality of cells in a network, thereby providing an intermediate connection to bridge any gap encountered during a hard-handover between the plurality of cells.

In other words the above description provides a communication network configured to form a plurality of sector sub-cells each covered by one emitter, where each sector sub-cell is hard constrained by the opening angle of a sector emitter in the base station (access point) and soft constrained by the range of its emissions. The hard constraints form a sharp cut-off of the emission power, to facilitate a hard handover to a neighbouring sector from the same base station. The soft constraint is characterized by a gradual decline of the received power from said sector cell, and a gradual increase of the signal strength from other sector sub-cells, to facilitate a soft, gradual handover from one sub-cell emitter, to a another sub-cell emitter on another base station. Hard handovers are most likely to occur between sub-cells of the same base station, and client device transitions from one base station to another base station are more unlikely (under normal operation these transitions would be handle by soft handover at PHY layer).

Figure 9 shows a side-view of a one dimensional cell comprising two ceiling based transceivers 901 and 902 and a client-device 903 comprising a multi-directional receiver placed on a table 910. The ceiling-based transceiver 901 emits modulated light within a solid angle demarcated by the lines 901-1 and 901-2, the light coverage area includes part of the floor 904 and of the table 903. Likewise the transceiver 902 emits light within the solid angle demarcated by the lines 902-1 and 902-2.

The client device 903 is positioned on the table and comprises a multi directional transceiver as well as a MIMO modem. The client device may comprise additional functionality and may be e.g. a laptop, or a dedicated optical wireless

communication receiver. The client device, may include a processor, memory and/or storage and may be arranged to provide services such as web-browsing, audio-playback, video playback, or even general computing

The multidirectional transceiver may comprise a single transceiver module with multiple sensors, or alternatively may comprise multiple transceiver modules, each with a single sensor. One of the directional transceivers/sensors of the client device senses light arriving at the device 903 from the left quadrant, indicated by the hatched area 907, the other directional transceiver/sensor of the client device senses light arriving at the device 903 from the right quadrant indicated by the hatched area 908. This separation in this case results from each sensor covering one of the quadrants formed by the table 910 top and the plane 905 orthogonal to the image.

Although the client device in this case has two light sensitive regions indicated by the hatched areas 907 and 908, it will be clear to those of even ordinary skill that this is not an essential limitation.

The directionality of the sensor, in case of optical wireless communication may be realized using optical means know to those skilled in the art, such as collimators, reflectors, shielding and/or lenses. As a result of the directional sensitivity the respective transceivers substantially receive signal from one of the two ceiling mounted transceivers 901, 902.

Figure 10A shows a plan- view of a two-dimensional network comprising four transceivers 1001, 1002, 1003, 1004, that in this example will be used as transmitters of signals. The transceivers coverage area is indicated by the corresponding circle segment 1001a, 1002a, 1003a, 1004a. Further a multi-directional transceiver 1010 is provided in the centre that here will be described as a receiver of signals.

The multi-directional transceiver 1010 in this example is a four-direction transceiver, that is aligned such that the respective sensor (and/or transceiver) can receive signals from the quadrant Q1. In this example it is clear that the respective sensors that sense Ql, Q2, Q3 and Q4 receive signals transmitted by the four transceivers 1001, 1002, 1003, 1004.

Figure 10B shows a plan- view of a two-dimensional network comprising four transmitters/transceivers in a rectangular cell and one poorly aligned client-device lOlOr comprising a multi-directional receiver sensing in 4-quadrants. As a result of the rotation of the client device relative to the transceivers 1001, 1002, 1003 and 1004, the sensors that sense Q3 and Q2 will be the only sensors that actually sense a signal from the transceivers. In this case the MIMO modem will need to derive the distinct signal contributions from the sensor signals out of the output of the quadrant Q2 and Q3 sensors.

Figure 10B illustrates that it may be beneficial to use a multi-directional receiver that has more sensors than transmitters; as due to the particularities of the cell layout, some sensors may not sense transmitter signals, in fact in this case it would have been advantageous to use the client device 1013 as presented in Figure IOC.

Figure IOC shows a plan- view of a two-dimensional network comprising four transmitters/transceivers in a rectangular cell and one client-device comprising a multi directional receiver sensing with two sensors sensing opposite halves HI and H2 of the space around it.

As exemplified by this example the sensors each cover 180 degrees, together complementary covering the 360 degrees around the device. As a result, each sensor captures two of the transmitting transceivers 1001 and these signals will need to be provided to the MIMO modem. In practice all signal - carrying signals from the sensors (e.g. above a threshold level) will need to be relayed to the MIMO modem for further processing. Figure IOC further depicts a number of alternate client-device transceiver directional-sensor configurations, one with eight sensing regions 1013, and one with three sensing regions 1014. Both could be applied for example in the situation of Figure IOC to better isolate the respective transmitting transceivers.

Although the above examples focus on a situation wherein a stationary client- device is being used in conjunction with a multi-directional transceiver, client-devices are not necessarily limited to stationary client-devices. In addition, it should be noted that although the examples are one or two-dimensional, in practice a system will operate in three- dimensions, which implies that e.g. in case of a light sensor, the apertures for the respective sensors need to be devised to capture, the ceiling mounted transceivers of a typical cell.

Finally it should be noted that when the client-devices are not stationary and when such devices may be tilted or yawed; then additional sensors may be leveraged to compensate for tilt or yaw.

The concepts described herein above apply to a downlink, thus from a plurality of base stations (e.g. at the ceiling) to a user client device. The same cell concepts also apply for the uplink, from the client device to the base station. The system may use the concept described herein for both downlink and uplink configurations.

For example, the plurality of emitters (or transmitters) can be located at the user client device, e.g. as a sectorized emitter, while receivers are located at the (multiple) base stations at the comers of the polygonal cell. Signals received at these base stations are then sent to the common modem.

It is further noted that MIMO modems, for modulating, demodulating OFDM or CDMA signals are well know in the art of telecommunications. MIMO modems may be implemented purely in hardware using e.g. an FPGA, or a dedicated ASIC. However, more and more silicon-on-chip solutions are provided wherein part of the MIMO functionality; in particular the real-time and high-throughput elements are implemented in hardware, whereas the higher level, protocols are handled by means of micro-controllers and/or processors.

It will be appreciated that the above embodiments have been described only by way of example. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims.

In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a” or "an" does not exclude a plurality. A single processor or other unit may fulfil the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. A computer program may be stored and/or distributed on a suitable medium, such as an optical storage medium or a solid- state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems. Any reference signs in the claims should not be construed as limiting the scope.

Claims

CLAIMS:

1. A communications network (100) comprising a plurality of cells (6,7), each cell having a polygonal coverage area (102) in a top-down 2D projection in a horizontal plane, the cell arranged to provide a channel for transmitting data to one or more client devices (108) within the coverage area; the system comprising, for each cell:

a plurality of emitters (104), each emitter positioned at a different respective comer of the polygonal cell, and configured to emit a respective beam (106) of radiation directed inwards with a footprint in said plane defining edges of the cell meeting at the respective comer to provide a respective coverage sub-area (106) so that the sub-areas combine to extend over the coverage area of the cell; and

a modem (112) associated with a separate cell wherein the modem is connected via a respective branch (110) to each of the emitters of the associated cell, and being thereby operable to modulate at least part of the data into the radiation of each of the respective beams of the associated cell,

wherein a first modem of a first cell of the communications network is configured to:

perform said modulation using a modulation method for simultaneous multibeam reception, and to send different parts of the data simultaneously over the channel, via said respective beams of the multiple emitters from different respective directions, such that the simultaneous beams are received at the client device along multiple respective paths from different respective angles,

perform a hard handover for a client device associated with the first modem to a second modem, when the client device moves from a first cell associated with the first modem to a second cell associated with the second modem and

perform a soft handover of the client device associated with the first modem when within the first cell, by adapting a proportion of the data transmitted to the client device via respective sub-channels of respective emitters of the same cell, such that the proportions of the data emitted are varied.

2. The network of claim 1, wherein the modem is configured to perform said modulation using orthogonal frequency division multiplexing.

3. The network of any proceeding claim, wherein the radiation comprises: light or line-of-sight high frequency radio.

4. The network of any proceeding claim, wherein the radiation is radio, and each emitter is configured to emit the respective beam by beamforming using an array of radio emitting elements.

5. The network of any proceeding claim, wherein the radiation is light, and each emitter is configured to emit the respective beam of light by beamforming using an array of light emitting elements.

6. The network of any proceeding claim, wherein the radiation is light, and each emitter is configured to emit the respective beam of light with distribution controlled by one or more optical components.

7. The network of claim 1, wherein the hard-handover is performed at the cell edge only.

8. The network of any preceding claim, comprising a filler cell (702), the filler cell having its own modem (706) and being located between some or all of the plurality of cells (102a, 102b), thereby providing an intermediate connection to bridge any gap between the plurality of cells during a hard-handover.

9. A method of forming a communication network (100) comprising a plurality of cells (6,7), each cell having a polygonal coverage area (102) in a top-down 2D projection in a horizontal plane, the cell arranged to provide a channel for transmitting data to one or more client devices (108) within the coverage area; the method comprising, for forming each cell:

positioning each emitter (104) at a different respective comer of the polygonal cell to emit a respective beam (106) of radiation directed inwards with a footprint in said plane defining edges of the cell meeting at the respective comer to provide a respective coverage sub-area (106) so that the sub-areas combine to extend over the coverage area of the cell; and connecting a modem (112) associated with a cell to each of the emitters of the associated cell via a respective branch (110) of the modem, and being thereby operable to modulate at least part of the data into the radiation of each of the respective beamsof the associated cell

the method comprising a first modem of a first cell of the communications network:

performing said modulation using a modulation method for simultaneous multibeam reception, and to send different parts of the data simultaneously over the channel, via said respective beams of the multiple emitters from different respective directions, such that the simultaneous beams are received at the client device along multiple respective paths from different respective angles,

performing a hard handover for a client device associated with the first modem to a second modem, when the client device moves from a first cell associated with the first modem to a second cell associated with the second modem and

performing a soft handover of the client device associated with the first modem when within the first cell, by adapting a proportion of the data transmitted to the client device via respective sub-channels of respective emitters of the same cell, such that the proportions of the data emitted are varied.

10. A modem for use in a communications network (100) comprising a plurality of cells (6,7), each cell having a polygonal coverage area (102) in a top-down 2D projection in a horizontal plane, the cell arranged to provide a channel for transmitting data to one or more client devices (108) within the coverage area; the system comprising, for each cell:

a plurality of emitters (104), each emitter positioned at a different respective comer of the polygonal cell, and configured to emit a respective beam (106) of radiation directed inwards with a footprint in said plane defining edges of the cell meeting at the respective comer to provide a respective coverage sub-area (106) so that the sub-areas combine to extend over the coverage area of the cell,

the modem (112) for association with a separate cell wherein the modem is connected via a respective branch (110) to each of the emitters of the associated cell, and being thereby operable to modulate at least part of the data into the radiation of each of the respective beams of the associated cell,

wherein the modem when associated with a first cell of the communications network is configured to:

perform a soft handover of the client device associated with the first modem when within the first cell, by adapting a proportion of the data transmitted to the client device via respective sub-channels of respective emitters of the same cell, such that the proportions of the data emitted are varied .

11. A kit of parts comprising a base station (400) and a plurality of modems according to claim 10, for use in a communications network (100) comprising a plurality of cells (6,7), each cell having a polygonal coverage area (102) in a top-down 2D projection in a horizontal plane, each cell arranged to provide a channel for transmitting data to one or more client devices (108) within the coverage area; the base station comprising:

a plurality of emitters (1, 2, 3, 4), each emitter configured to emit a respective beam (402) of radiation to provide a respective coverage sub-area (402) of a different respective one of said cells, the sub-area constrained by an opening angle of the emitter such that the sub-areas combine to extend around the circumference of the base station in a horizontal plane, each emitter configured to receive a signal from a different modem (602a, 602b) for outputting to a different cell; and

wherein the base station is configured to enable each of the emitters to be controlled by a different modem out of the plurality of modems.