TWI556589B - Cdma/o-cdma with code-shifting technique and system thereof - Google Patents

Description

Code division multiplexing transmission method and system using non-integer displacement

The invention is a code division multiplexing assignment method, in particular to a code division method and system thereof for a code division multiplexing technology (CDMA/O-CDMA).

CDMA (Code Division Multiple Access) technology has the characteristics of good confidentiality, strong anti-interference ability and high system capacity, and is widely used in a new generation of mobile communication technologies. The principle of optical code division multiple access (O-CDMA) is similar to that of traditional wireless communication. It has effective information transmission, dynamic bandwidth assignment, asynchronous access, and multiple users. At the same time, using the same channel and other characteristics, O-CDMA is widely used in high-speed optical fiber communication, infrared wireless communication and so on.

Although CDMA/O-CDMA has been tested and tested in the market and its performance has certain qualities, the transmission quality of this technology has long been limited by the dilemma between code length, code weight and the number of channels used. O-CDMA coding has a great impact on system performance, multiple-access interference (MAI), error rate, and the like. Choosing a relatively long code is not easy for system design and hardware selection. Therefore, improving the current coding mode and proposing a more efficient coding method under limited hardware conditions can help the development of CDMA and O-CDMA. beneficial.

In order to solve the technical problems related to code length, bandwidth, error probability and the like in the prior art, the present invention proposes to use different degrees of time translation or delay (or no displacement) means before transmission, and the translation or delay system It is a non-integer, so that each transmitted code maintains independence and improves the transmission quality. Thus, the code used in the present invention can achieve better transmission quality under limited length, weight and hardware conditions. There are technical problems.

The invention provides a code division multiplexing transmission method using non-integer time shift, the steps of which include:

The code is continuously generated by an encoding module, each code comprising a plurality of time units, each code comprising more than one pulse;

The pulse of at least one of the at least one code moves in a non-integer manner of movement of the time unit; and

All codes are output to a receiving end in one channel.

Wherein, the code is a one-dimensional code or a two-dimensional code.

Wherein, the displacement of the pulses in each code may be lead or delayed displacement or no displacement.

Wherein, the displacement of each pulse in the code is asynchronous.

Among them, the code displacement uses a cyclic displacement.

The present invention further provides a code division multiplexing transmission system using a non-integer shift, which comprises a transmitting end, a receiving end and a channel respectively transmitting a connection with the transmitting end and the receiving end signal, wherein the transmitting end continuously generates a channel Dimensional or two-dimensional code, each code comprising a plurality of time units and pulses present in the time unit, at least one of the at least one code being shifted or delayed by a non-integer unit, the transmitting end of the code being lightwave , wireless or infrared wireless form output to the receiving end.

Wherein, the delay or shift of each pulse in the code is asynchronous.

Wherein, the displacement or delay of each pulse in the code is random, and the displacement or delay is less than one such time unit.

Based on the foregoing description, the present invention has the following features:

1. Regardless of whether a one-dimensional or two-dimensional code is used, the present invention is equally applicable.

2. Propose a random, non-integer (or zero) displacement method that allows the system to achieve a lower probability of error under limited hardware conditions.

3. When choosing a relatively large g (number of sub-unit cuts), the probability of error can be significantly reduced.

The system of the present invention using a non-integer (including zero) time-shifted code division multiplexing transmission method may include a transmitter, a receiver, and one of a signal transmission connection with the receiver and the receiver, respectively. aisle. The transmitting end is based on the requirements of optical code division multiplexing (O-CDMA), and may include an encoding module (encoder) to generate one-dimensional using a one-dimensional (1D) encoding module or a two-dimensional (2D) encoding module. Or a two-dimensional code, wherein the technology used in the coding module is an existing coding technology related to CDMA technology, and details are not described herein again.

Each codeword contains a plurality of time units (chips or time slots), and each time unit can be cut into a plurality of equal sub-chips or sub-slots, where g is ≥ 1, which represents The granularity of the code. The pulse of the time unit within the code can be moved (or not moved) randomly, independently, non-integer, because the movement of the pulse is random, and the cross-correlation property of the O-CDMA code is thus changed. As a function of g, it has been confirmed that the finer the fineness (g is larger), the more likely it is to represent the possible position of the pulse, thus reducing the collision of the code transmission process and the multiple access interference (MAI) caused by the user's proliferation. problem. The correlation analysis pulse is detailed later.

The following is a description of the two-dimensional code. The non-integer (including zero) time-shifted code division multiplexing transmission method steps of the present invention include:

1. Select a two-dimensional code, which is an LxN two-bit matrix containing L sets of wavelengths, N sets of time units, w pulses (weights, weights), and maximum cross-correlation values (λ _c ), where each code The w pulses within can be represented as w ordered pairs: C _i = [(λ ₀ , t _{λ 0} ), (λ ₁ , t _{λ 1} ), . . . , (λ _w-1 , t _{λ w − 1} )], where each ordered pair represents a pulse of wavelength λ _j at t _j ∈ {0, 1, . . , N-1} for j ∈ {0, 1, . . ,w−1}. For a one-dimensional array, it is expressed as λ ₀ = λ ₁ = . . . = λ _w-1 .

2. Select the number of sub-units g, which is a positive integer greater than 1, such that a plurality of equal sub-units are generated in each single-time unit, and each pulse is randomly and independently advanced in each sub-unit within the time unit in which it is located. Or delay moving or not moving. The so-called random movement means that the movement of each pulse in time units is a random delay or a number of sub-units of advance movement; the so-called independent movement refers to the pulse movement of different time units.

3. Add a random sub-slot-shift value to the position of the original time unit corresponding to the pulse, used to specify a new time unit plus sub-unit after the move (time-slot-plus- Sub-slot), such a random displacement can be considered as a time delay. Assuming that the width of the time unit is 1, such a subunit has a width of 1/g, so a time-delay factor can be defined:

τ _{g,j,s j} ∈ {0, 1/g, 2/g, . . . , (g−1)/g}, which represents several sub-units of the jth pulse delay (one-dimensional code), or represents the first The delay of the pulse of the jth wavelength (2-dimensional code), where j ∈ [0 , w−1]. Sj is a random integer selected from {0, 1, 2, . . . , g − 1}, the subunit corresponding to the beginning of the jth pulse. The sequence pair after displacement can be expressed as: [(λ ₀ , t _{λ 0} +τ _{g,0,s 0} ), (λ ₁ , t _{λ 1} +τ _{g,1,s 1} ), . . . , (λ _{w- 1} , t _{λ w − 1} +τ _{g, w − 1,sw − 1} )], where τ _{g,0,s 0} ,...τ _{g, w − 1, and sw − 1} are randomly selected from { 0,1 /g,2/g, . . . ,(g−1)/g}, and the displacement or delay of the code can be non-uniform.

Please refer to FIG. 1 , which is a schematic example of a non-integer-shifted one-dimensional code according to the embodiment. The data content of the code is 010010000001, and g is 2 (representing 2 sub-units per time unit), because the weight w is 3. There are three sets of random delays τ _{2,0, s0} , τ _{2,1, s1} , τ _{2,2, s2} are selected from the set of {0, 1/2}, where sj ∈ {0, 1}, J∈[0,2]. Figure 1(a) _{shows an} example of a pulse without displacement. Figure 1(b) _{shows the} displacement of τ _2,0,1 =1/2, τ _2,1,1 =1/2, τ _2,2,0 =0. Figure 1(c) _{shows the displacement} of τ _2,0,0 =0, τ _2,1,0 =0, τ _2,2,1 = 1/2. Half of the last pulse in Figure 1(c) has been moved to the first subunit of the first time unit because the displacement can be a cyclic technique.

In order to prove that the non-integer displacement proposed in this embodiment has substantial benefits for CDMA or O-CDMA, the performance of the post-displacement code can be analyzed, and the error probability of the hardware can be analyzed. The error probability of the OOK O-CDMA system can be the following. formula:

Among them, 1/2 comes from the transmission rate of two bits, K represents the number of simultaneous users, Qλc, j represents the probability of j collisions of the decoder used by O-CDMA system, where j ∈ [0 , λc], Where λc>=1 is the maximum cross-correlation value for this code, where:

Please refer to Figures 2 and 3 to simulate the error probability of w = L = {3, 5, 7}, N = {9, 25, 49}, λc=1 and g = {2, 3, 4}. The result is shown in Fig. 2. It shows that when the number of simultaneous users is fixed, selecting a relatively large g (having more subunits) has a lower probability of false performance. Figure 3 shows more significantly that when the selected g is larger, the probability of error can be significantly reduced.

Based on the foregoing description, the present invention has the following features:

1. Regardless of whether a one-dimensional or two-dimensional code is used, the present invention is equally applicable.

2. A random, non-integer displacement method is proposed to allow the system to achieve a lower probability of error under limited hardware conditions.

3. When choosing a relatively large g, you can significantly reduce the probability of error.

FIG. 1 is an example of a non-integral digital shift according to an embodiment of the present invention. 2 and 3 are diagrams showing an error probability simulation analysis of an embodiment of the present invention.

Claims (8)

A code division multiplexing transmission method using non-integer shifts, the method comprising: continuously generating codes by an encoding module, each code comprising a plurality of time units, each code comprising a plurality of pulses; the complex pulse of at least one code The movement mode of the time unit of the non-integer, wherein the complex pulse of the at least one code has a plurality of random delays; and all the codes are outputted to a receiving end by one channel. For example, the first aspect of the patent application scope uses a non-integer shifting code division multiplexing transmission method, wherein the code is a one-dimensional code or a two-dimensional code. If the first or second application of the patent application uses a non-integer displacement code division multiplexing transmission method, the displacement of the pulse in each code may be lead or delayed displacement or no movement. For example, in the third application of the patent application, a code division multiplexing transmission method using a non-integer shift is used, and the displacement of each pulse in the code is asynchronous. For example, the fourth aspect of the patent application uses a non-integer displacement code division multiplexing transmission method, wherein the displacement of the code uses a cyclic displacement. A code division multiplexing transmission system using a non-integer shift, comprising a transmitting end, a receiving end and a channel respectively transmitting a connection with the transmitting end and the receiving end signal, wherein the transmitting end continuously generates one-dimensional or two-dimensional a code, each code comprising a plurality of time units and a pulse present in the time unit, at least one of the at least one code being shifted or delayed by the time unit of a non-integer, the transmitting end of the code being light wave, The wireless or infrared wireless form is output to the receiving end; wherein the complex pulse of the at least one code has a plurality of random delays. As with the code division multiplexing transmission system using non-integer displacement as described in claim 6, the delay or shift of each pulse in the code is asynchronous or non-displacement. As with the code division multiplexing transmission system using non-integer displacement as described in claim 6 or 7, the displacement or delay of each pulse in the code is random, and the displacement or delay is less than one such time unit.