AU2007209824A1

AU2007209824A1 - Improved method and system for mapping traffic predictions with respect to telematics and route guidance applications

Info

Publication number: AU2007209824A1
Application number: AU2007209824A
Authority: AU
Inventors: Yosef Mintz
Original assignee: Individual
Current assignee: Individual
Priority date: 2001-02-09
Filing date: 2007-08-17
Publication date: 2007-09-06

Description

-1-

AUSTRALIA

PATENTS ACT 1990 COMPLETE SPECIFICATION FOR A STANDARD PATENT

ORIGINAL

Name of Applicant/s: Actual Inventor/s: Address for Service is: Yosef Mintz Yosef Mintz SHELSTON IP Margaret Street SYDNEY NSW 2000 CCN: 3710000352 Attorney Code: SW Telephone No: Facsimile No.

(02) 9777 1111 (02) 9241 4666 Invention Title: IMPROVED METHOD AND SYSTEM FOR MAPPING TRAFFIC PREDICTIONS WITH RESPECT TO TELEMATICS AND ROUTE GUIDANCE APPLICATIONS Details of Original Application No. 2002256855 dated 08 Feb 2002 The following statement is a full description of this invention, including the best method of performing it known to me/us:- File: 40379AUP01 501273937 1 OOC/5844 -2- O IMPROVED METHOD AND SYSTEM FOR MAPPING TRAFFIC PREDICTIONS WITH C RESPECT TO TELEMATICS AND ROUTE GUIDANCE APPLICATIONS ;The present invention is a divisional application of Australian Patent Application No. 2002256855, whose entire contents are hereby incorporated by reference.

Field of the Invention (Ni This invention relates generally to a method and system for mapping potential 00 traffic loads in forward time intervals, according to various criteria which might indicate i erratic traffic, as a result of expected increase in the number of Mobile Telematics Units S(MTU) and In-Car Navigation Systems (CNS) users that use Dynamic Route Guidance S 10 (DRG). In particular, the method and system aims to provide an efficient means to estimate the potential increase or decrease in the number of vehicles in selected places (inconsistent traffic load), by using a radio system, in order to help in determining levels of a potential erratic behavior in the traffic due to the use of DRG by a significant percentage of vehicles. This system and method may further help to investigate sources of causes of erratic traffic and their level of effect, including the use of traffic information and reactions of drivers to telematics applications. This could help to improve traffic predictions for the use of traffic control and DRG. In particular, this method provides the ability to make use of a mapping system platform which has the capability to allocate preassigned slots or groups of slots for the detection of signal responses from mobiles that have probe response capability. The above identified system is mainly characterized by the ability of the mobiles to select time/frequency slots for response signals according to a mapping system query and according to a predetermined protocol. The detection of mobile transmission signals is mainly characterized by energy detection of mobile transmitted signals in allocated slots and hence there is no need for a repeat in mobile transmission as a result of signal collisions in the same slot. The non mobile platform of such a mapping system, which may be referred to hereinafter as Slot Oriented Discrimination Mapping System (SODMS), or as otherwise referred to, as well as the mobile (probe) response capability are described in US applications 09/945,257 and 09/998,061 filed November 30, 2001 and PCT/IB00/00239 and their own references.

Description of Related Art Any discussion of the prior art throughout the specification should in no way be considered as an admission that such prior art is widely known or forms part of common general knowledge in the field.

For example PCT publication WO 96/14586, published 17 May 1996, the N disclosure of which is incorporated herein by reference, describes, inter alia, a system for ;mapping of vehicles in congestion. In one embodiment applicable to the mapping system platform, described in the above publication, a central station broadcasts a call to the vehicles which requests for example those vehicles which are stopped or which have an average velocity below a given value to broadcast a signal indicative of their position.

Such signals are broadcast in slots, each of which represent one bit (yes or no) which (Ni 00 relates to a position. Preferably, only one logical slot (that may be represented by more Sthan one actual slot) is used to define the related position. Such signals are then used to (Ni generate a map of those regions for which traffic is delayed or otherwise moving slowly.

In the above-identified prior art, the possible construction of consistent traffic database for possible use with traffic predictions have been described. Such database could be constructed by traffic mapping of queues, when quasi-stationary (temporary stationary) statistics of traffic flow in a mapped road, at certain periods of time of a day, and for days in which traffic conditions, are considered to be repetitive. Such collected information, average arrival rates, could be used as off line database to predict traffic in conjunction with real time updates of mapped queues using statistical methods known in the art. By using the mapping method in this embodiment for mapping the potential effects of erratic traffic, either when produced as part of the current traffic mapping application of the mapping system platform (described by the above identified prior art) or by a separate platform with similar communication capabilities, it is possible to update the consistent traffic database by incorporating inconsistent traffic predictions.

Background of the Invention The expected increase in the number of Telematics applications by MTUs used with off-board or on-board route guidance as well as the increase in the number of CNS users would increase the percentage of vehicles that would use Dynamic Route Guidance and would hence result in unpredicted changes in traffic load which has the potential to cause erratic traffic.

Traditional traffic predictions could use a database of consistent traffic in order to predict traffic according to expected traffic loads, possibly also according to prior knowledge about the behavior of the traffic and the current conditions of traffic. However DRG effects on traffic might mostly be unpredictable by such a database.

This could be the result even though there is a priori information about off board N DRG (routs plans provided by common service centers), since deviations in the schedule ;of routes and possible use of alternative routes could in a short time make prior knowledge to become irrelevant to traffic prediction. Thus it would be valuable to have a means to update a traffic database that would be used in conjunction with consistent traffic information and possibly with other prior knowledge including current traffic information in order to improve the capability to predict potential changes in traffic.

(,i 00 0Consistent Traffic is defined as such traffic that has a repetitive characteristic, N with respect to specific time periods and places, certain hour in a certain day of the week in a certain road). Consistent Traffic is a result of behavior patterns that from a statistical point of view usually and in general may be characterized. Such traffic characteristics may be stored in an off-line data base which may contribute to traffic predictions.

Inconsistent Traffic is defined as such traffic that has a non repetitive and erratic characteristic with respect to specific time periods and places. Such traffic may for example be the result of the ability by the individual driver to change routes according to current traffic loads. As the number of drivers that have access to detailed information on currently changing traffic increases, and as the number of drivers that possess in-car sophisticated capability to individually vary their previous route plans, and the less coordination if any exists amongst various drivers, the more inconsistent would become such traffic. Inconsistent Traffic is difficult if at all possible to be characterized on a statistical basis. Such traffic tends to be in general unpredictable, and leads to unpredictable traffic loads.

The inconsistent traffic is expected to become a significant issue in the control of the traffic when a significant percentage of cars will be using dynamic route guidance and as a result might probably, in themselves cause unexpected traffic loads at certain places that would affect the traffic and reduce the efficiency of dynamic route guidance.

Traffic information used with Dynamic Route Guidance (DRG) could be one reason for the inconsistency in the traffic due to changes in planned routes, while driver preferences, deviation from schedule, or reaction to local based services could be other causes for an inconsistency in the conditions of the traffic.

One general approach to resolve the problem of predicting inconsistent traffic is to centralize the control of the individual driver routes. This is not the approach which is considered in the following embodiment of the invention as it leads to centralized DRG which has many disadvantages beside feasibility problems with large scale implementation.

As further explained, apart from the contribution of traffic predictions of inconsistent traffic to traffic control the predictions could further lead to a relatively low cost implementation of an anonymous predictive DRG approach based on distributed intelligence of the in car computers and also to contribute to the implementation of more 00 efficient telematics applications.

(Ni Predictions for inconsistent traffic is based on a process of traffic load estimation for predetermined place and time interval, (for example, estimating the number of vehicles that use in-car navigation computers which are expected to pass in a certain road in a certain forward time interval). However when the source of such information is limited to car navigation units that use dynamic route guidance only, and the estimation process is the only means for such predictions, it would be required that most of the cars should use car navigation systems. In practice such a situation would doubtfully be viable. However, the situation when a significant percentage of vehicular systems would most probably be using Dynamic Route Guidance (DRG) may be considered realistic in the not too distant future, and hence inconsistent traffic would begin to appear at an early stage, whereas reliable traffic prediction for this situation would not yet be available. With the lack of traffic predictions, the problems that would be encountered at such stages could lead to a significant dilemma by the individual drivers, about the efficiency of Dynamic Route Guidance. The dilemma would be whether to consider recommended DRG according to current traffic, while ignoring unpredictable traffic that might result due to the significant number of DRG users, or ignoring the recommended DRG. For such early stages of inconsistent traffic the following embodiment suggests a modified method of traffic predictions in order to enable reliable prediction at such early stages. Traffic load predictions would preferably refer mostly to sensitive roads that encounter recurrent traffic jams.

It is an object of the present invention to overcome or ameliorate at least one of the disadvantages of the prior art, or to provide a useful alternative.

Summary of the Invention )According to a first aspect of the invention, there is provided a method of ;estimating a demand for an item, wherein the item is to be offered according to said demand, the method comprising: transmitting to mobile units information related to said item; and receiving a response generated by a mobile unit of said mobile units, wherein the (Ni response is generated according to a match process between the information related to

OO

1 said item as received by said mobile unit and information related to said item as stored in (K a list of items in said mobile unit.

Unless the context clearly requires otherwise, throughout the description and the claims, the words "comprise", "comprising", and the like are to be construed in an inclusive sense as opposed to an exclusive or exhaustive sense; that is to say, in the sense of "including, but not limited to".

According to a second aspect of the invention, there is provided a communication system for estimating a demand for an item, wherein the item is to be offered according to said demand, the system comprising: a transmitter to transmit to mobile units information related to said item; and a receiver to receive a response generated by a mobile unit of said mobile units, wherein the response is generated according to a match process between the information related to said item as received by said mobile unit and information related to said item as stored in a list of items in said mobile unit.

According to a third aspect of the invention, there is provided a method of controlling a number of responses from mobile units, wherein an item is to be offered to potential customers through said mobile units, according to one or more of the following criteria: limiting said responses to a predefined acceptable arrival range; limiting said responses according to a criterion of potential arrivals of vehicles carrying said mobile units from one or more directions; limiting said responses according to a criterion of potential arrivals of vehicles carrying said mobile units through one or more route segments; limiting said responses according to deviation in predicted traffic; and limiting said responses according to whether or not a responding mobile unit is expected to change one or more of its route segments at one or more forward time intervals according to a planned route in use by said mobile unit.

According to a further aspect of the invention there is provided a method of predicting load of traffic of vehicles that are travelling according to non reference route plan, provided with Dynamic Route Guidance capability of their PMMS, in a Forward N Time Interval related Route Segment and according to a predetermined protocol 00 0between mobile systems and a non mobile system platform of a SODMS, the method comprising: receiving by mobile units a traffic prediction query and according to a predetermined differential traffic load match process; performing a match process by each of the mobile units and, according to a match; enabling a predetermined response procedure wherein a response procedure in each mobile unit uses a predetermined random process to select an allocated slot in which to transmit a predetermined signal, which provides an improved way to predict traffic in conjunction with off line database statistics, preferably with such that are being adaptively corrected by prior data and method to predict traffic which do not include, or lack sufficient erratic traffic information.

According to a further aspect of the invention there is provided a method for estimating according to criteria and a predetermined protocol local demand according to SPL, preferably in conjunction with predicting respective load of differential traffic in forward time intervals for selected places which might result from a hunting trip application, and according to a further predetermined protocol between TCs and a non mobile system platform of a SODMS, the method comprising: receiving by TC units a query of a hunting trip application and according to a predetermined match process; performing a match process by each of the TC units and, according to a match; enabling a predetermined response procedure wherein a response procedure in each TC unit uses a predetermined random process to select an allocated slot in which to transmit a predetermined signal.

According to a further aspect of the invention there is provided a method of predicting load of traffic of vehicles that are travelling according to non reference route plan, provided with Dynamic Route Guidance capability of their PMMS, in a Forward N Time Interval related Route Segment and according to a predetermined protocol between mobile systems and a non mobile system platform of a SODMS, the method comprising: receiving by mobile units a traffic prediction query and according to a predetermined differential traffic load match process; performing a match process by each of the mobile units and, according to a 00 match; enabling a predetermined response procedure wherein a response S 10 procedure in each mobile unit uses a predetermined random process to select an allocated slot in which to transmit a predetermined signal, which provides an improved way to predict traffic in conjunction with off line database statistics., and wherein the database statistics are being adaptively corrected by prior data and method to predict traffic.

According to a further aspect of the invention there is provided a method for estimating according to criteria and a predetermined protocol local demand according to SPL, in conjunction with predicting respective load of differential traffic in forward time intervals for selected places which might result from a hunting trip application, and according to a further predetermined protocol between TCs and a non mobile system platform of a SODMS, the method comprising: receiving by TC units a query of a hunting trip application and according to a predetermined match process; performing a match process by each of the TC units and, according to a match; enabling a predetermined response procedure wherein a response procedure in each TC unit uses a predetermined random process to select an allocated slot in which to transmit a predetermined signal.

The present invention provides a preferred method and system for differential mapping of potential traffic loads in forward time intervals in selected places, which could be a result of DRG, in order to provide rapid and effective means for traffic prediction.

The mapping system, in which slots are allocated to probe responses, and mobile units that are equipped with route guidance with probe response capability in allocated slots, could be used as a platform for the following modified prediction method. The mobile unit would be referred to as Potential Mobile Mapping 0 System (PMMS). The route guidance capability of a PMMS could be based on either ;on board or off board route guidance. The prediction method described in the following could be implemented with such platforms, either with or without the implementation of the application of mapping of current traffic as part of this platform.

The non mobile part of the mapping system (non mobile systems), including the radio system and the mapping system, will be referred to as the non mobile system r platform. All applicable terms used in the above identified prior art, in connection with 00 0traffic mapping, and which are applicable and would contribute to the implementation O io of the following embodiment of the invention, will hold also for this application.

The aim of the differential mapping method for determining potential traffic Sloads is to update a traffic information database with information about deviation from rexpected traffic loads in forward time intervals for selected road segments in order to enable more accurate and prediction capability of the use of a traffic information database. Based on the inherent limitations of the database prediction capability (before deviation updates), prediction criteria are formulated and could be transmitted by means of the non mobile platform to the PMMS units. Such criteria are intended to enable the prediction of expected potential deviations from schedule and previously planned routes, at the level of the database requirements. The PMMS units could determine if they match the transmitted criteria, and if a match exists, would respond accordingly. This could also be considered as a method to improve accuracy levels of information in database that could help to predict traffic according to pre-investigation of local potential loads affected by DRG in selected forward time intervals. The level of basic information in such database could for example include consistent traffic, or higher level prediction capabilities.

For example, if the use of the database is based on prediction capabilities according to consistent traffic, then cars that change their planned route according to traffic information, most probably from the shortest route according to time and distance to one that most probably is shortest according to time, or other dynamic preference, could be used to indicate on possibly expected inconsistent traffic that is not taken into account within consistent traffic statistics. Thus it would be worth to first isolate this group of cars in order to estimate their contribution to the inconsistent traffic loads in specific road segments. Preferably, this information would then be taken into account in conjunction with a database of consistent traffic statistics, preferably updated with current real time updates of traffic, to determine current and predicted traffic information that would be currently updated accordingly. The isolation process would use prediction queries that would selectively target cars that made a change to their route or deviated from schedule, according to traffic tbinformation or other predetermined possible reasons such as a response of drivers to ;a telematics application. The queries determine the response criteria which will include but not be limited to the following a) vehicles that are planning to pass in a certain road at a certain forward time interval according to their modified route plan or schedule, and which did not plan to do so according to a reference route a default route or any other route that could be referred by the PMMS as a reference that may be determined according to criteria as part of a predetermined protocol), 00 and b) vehicles that did plan to pass in this road according to the reference route, 1 o and are not planning to do so according to the modified route plan or schedule at the above forward time interval.

Vehicles which are using their reference default) route will not respond to Squeries.

Criteria for determining whether a route is within reference conditions is default) or not, could be provided from a common external source, which considers the investigated level of possible effect on the traffic statistics. The reference default) route information may be formed either in the in-car (on board) systems, or received from external (off board) sources, and would preferably be determined by route plan and schedule. Thus, according to a predetermined protocol, a deviation in route or schedule would exclude tlhe route from being referred to as a reference route and would determine it to be a non reference route. The protocol would preferably include threshold levels of deviation.

Typical default routes are such which could be considered but not limited to conform with consistent traffic. Default routes could be determined according to common criteria the shortest route, preferably with time schedules), for mobile units participating in the following processes. Non default routes are such that have some significant effect on known traffic statistics as a result of deviation from schedule or from original route plan that could be considered as default routes.

The in-car system will incorporate a predetermined decision procedure, 3o described in the following.

In principle, a Differential Traffic Load Prediction (DTLP) process with respect to a Forward Time Interval related Route Segment (FTIRS refers to a time interval with respect to a route segment, usually a road segment) under investigation, could be implemented by means of two types of traffic prediction queries which would be transmitted by a mapping system to the PMMS units. The prediction queries include the prediction criteria, and are aimed at targeting groups of cars that are either expected to pass through the FTIRS under investigation and were not expected to do N so, according to database information, (non expected vehicles NEV), or are not texpected to pass through the FTIRS under investigation, and were expected to do ;so, according to the database information (expected vehicles -EV) Query A) type of a query with the aim of estimating the number of vehicles which on their reference route are not expected to pass through the investigated FTIRS, and on their non reference route are expected to pass through the investigated FTIRS, (non expected vehicles NEV), and Query B) type of a query with the aim of estimating the number of vehicles which 00 on their reference route are expected to pass through the investigated FTIRS and on io their non reference route are not expected to pass through the investigated FTIRS, (expected vehicles EV).

In order to enable responses in relation to forward time intervals, it is required Sthat the PMMS units would be equipped with the means of reference or mean to calculate reference to segments of planned routes and estimated travel time intervals along respective route segments. Preferably, an estimated time interval will be provided with respective confidence intervals.

Vehicles which are using a non reference planned route, will enable the response procedure according to the following decision procedure; If the received query is identified as Query A, then, according to the following differential traffic load match process result, if there is a match between FTIRS in the query and the planned non reference default) route (route in use), and there is no match between FTIRS in the query and the reference route, then enable the response procedure.

If the received query is identified as Query B, then, according to the following differential traffic load match process result, if there is a match between FTIRS in the query and their reference route, and there is no match between the FTIRS in the query and non reference route (route in use), then enable the response procedure.

Enabling the response procedure, in the predetermined decision procedure, would preferably be expanded to include additional criteria, for targeting vehicles. For example, with respect to Query A, additional criteria in checking an interval estimate for the probability to arrive within the investigated FTIRS, would preferably be taken into account as part of the decision procedure.

In order to alleviate the computation load in the in-car system, involved in frequent matching in response to above queries, it would be preferable to refer routes to predetermined area zones, and by a preliminary predetermined screening procedure, preceding the above decision procedure, vehicles whose planned (reference and non reference) routes do not cross area zones in which the FTIRS is N included, will not continue with the more detailed matching process in the above tb decision procedure.

;A number of communication slots will be preferably allocated for responders (cars which transmit in the allocated slots) in the response procedure, separately, with respect to each Query. Each of the targeted vehicles, (responders), in which the response procedure is enabled, will use a predetermined response procedure to select a slot in which to respond. This predetermined procedure would preferably use a uniformly distributed random selection of a slot out of all the allocated slots, to 00 transmit a signal.

S 10 In accordance with an embodiment of the invention, there is thus provided a method of predicting load of traffic of vehicles that are traveling according to non reference route plan, provided with Dynamic Route Guidance capability of their SPMMS, in a Forward Time Interval related Route Segment and according to a predetermined protocol between mobile systems and a non mobile system platform is of a SODMS, the method comprising: receiving by mobile units a traffic prediction query and according to a predetermined differential traffic load match process, performing a match process by each of the mobile units and, according to a match, enabling a predetermined response procedure wherein a response procedure in each mobile unit uses a predetermined random process to select an allocated slot in which to transmit a predetermined signal, which provides an improved way to predict traffic in conjunction with off line database statistics, preferably with such that are being adaptively corrected by prior data and method to predict traffic which do not include, or lack sufficient erratic traffic information.

In another embodiment of the invention it would be valuable to use traffic predictions in conjunction with applications which have a potential to cause erratic conditions of traffic. Such applications could include local based services in telematics and in particular position related commerce (p-commerce sometimes referred to as I-commerce or m-commerce). There might be different ways to implement p-commerce and hence to increase the level of unpredicted traffic. For example in order to improve p-commerce applications, it would be an advantage to large stock holders and others to have a query tool that would help them to identify sufficient demand, preferably according to prices and including non solicited products, for special offers. This could create a hunting trip environment. With such a tool, queries could be provided in a way similar to an auction process, preferably by a broadcast message to the telematics users, with respect to products with possibly (N one or more ranges of prices. The user, usually a driver, will have a stored list of tpreferences for products, in his Telematics Computer (TC which could be the computer of a Telematics-PMMS) that would be matched with broadcast messages according to preferences in the list. For example, a stored product list (SPL) which may include products with ranges of prices could enable the TC to respond to a broadcast query. If such responses would provide information about the estimated number of the potential clients and possibly their position distribution it would enable the vendor to determine a time window and price for a special offer according to 00 demand. The offer could then target the potential clients. Most probably this would io target the responders who would contribute to the decision making. When considering a system platform with capabilities such as suggested for a traffic mapping system, both, in this embodiments of the invention and in the reference prior art, together with telematics mobile unit with PMMS capabilities, which enable to estimate the number of responders to a query by random response in pre determined number of slots, it would be possible to implement a hunting trip application, efficiently.

Thus in accordance with this embodiment of the invention, there is thus provided a method for estimating according to criteria and a predetermined protocol local demand for products or services) according to SPL, preferably in conjunction with predicting respective load of differential traffic in forward time intervals for selected places which might result from a hunting trip application, and according to a further predetermined protocol between TCs and a non mobile system platform of a SODMS, the method comprising: receiving by TC units a query of a hunting trip application and according to a predetermined match process, performing a match process by each of the TC units and, according to a match, enabling a predetermined response procedure wherein a response procedure in each TC unit uses a predetermined random process to select an allocated slot in which to transmit a predetermined signal.

BRIEF DESCRIPTION OF THE DRAWINGS Fig 1, describes an iterative estimation procedure that is preferably used with more than a single iteration of estimation (separate allocation of slots with each iteration). The iterative estimation procedure is preferably aimed to obtain an estimated result of the number of responders with a restricted acceptable error level and to reduce biasness. The error level of the estimate in a single iteration is a N, function of the ratio between the number of slots in which responses are detected tb(responding slots) and the given number of allocated slots. Since the ratio of responding slots to a given number of allocated slots would be a result of the number of responders, it is desirable to assess in advance a realistic anticipated range of s responders, in order to determine a minimal number of initial allocated slots.

DETAILED DESCRIPTION OF THE DRAWINGS Fig 1, describes an iterative estimation procedure that is preferably used with 00 more than a single iteration of estimation (separate allocation of slots provided with i o each performed iteration). The iterative estimation procedure is preferably aimed to obtain an estimated result of the number of responders with a restricted acceptable error level, to reduce biasness and to check consistency. The error level of the ri estimate in a single iteration is a function of the ratio between the number of slots in which responses are detected (responding slots) and the given number of allocated slots. Since the ratio of responding slots to a given number of allocated slots would be a result of the number of responders, it is desirable to assess in advance a realistic anticipated range of responders, in order to determine a minimal number of initial allocated slots. Since such realistic ranges of responders could be anticipated from statistical data, according to time and place, then a data base of possible initial ranges would preferably be evolved for any particular urban entity, preferably as probability distribution from which ranges of confidence intervals could be derived.

Combined estimates that can use joint probabilities and Bayesian methods as described above with respect to Fig. 1 are described in more detail in the detailed description of Preferred Embodiment of the invention.

DETAILED DESCRIPTION OF A PREFERRED EMBODIMENT The present invention provides a preferred method and system for differential mapping of potential traffic loads in forward time intervals in selected places, which could be a result of DRG, in order to provide rapid and effective means for traffic prediction. The mapping system, in which slots are allocated to probe responses, and mobile units that are equipped with route guidance with probe response capability in allocated slots, could be used as a platform for the following modified prediction method. The mobile unit would be referred to as Potential Mobile Mapping System (PMMS). The route guidance capability of a PMMS could be based on either on board or off board route guidance. The prediction method described in the following could be implemented with such platforms, either with or without the implementation of the application of mapping of current traffic as part of this platform.

N The non mobile part of the mapping system (non mobile systems), including the radio tbsystem and the mapping system, will be referred to as the non mobile system ;platform. All applicable terms used in the above identified prior art, in connection with traffic mapping, and which are applicable and would contribute to the implementation of the following embodiment of the invention, will hold also for this application.

The aim of the differential mapping method for determining potential traffic loads is to update a traffic information database with information about deviation from 0 expected traffic loads in forward time intervals for selected road segments in order to 00 enable more accurate and prediction capability of the use of a traffic information S 10 database. Based on the inherent limitations of the database prediction capability (before deviation updates), prediction criteria are formulated and could be transmitted by means of the non mobile platform to the PMMS units. Such criteria are C intended to enable the prediction of expected potential deviations from schedule and previously planned routes, at the level of the database requirements. The PMMS is units could determine if they match the transmitted criteria, and if a match exists, would respond accordingly. This could also be considered as a method to improve accuracy levels of information in database that could help to predict traffic according to pre-investigation of local potential loads affected by DRG in selected forward time intervals. The level of basic information in such database could for example include consistent traffic, or higher level prediction capabilities.

For example, if the use of the database is based on prediction capabilities according to consistent traffic, then cars that change their planned route according to traffic information, most probably from the shortest route according to time and distance to one that most probably is shortest according to time, or other dynamic preference, could be used to indicate on possibly expected inconsistent traffic that is not taken into account within consistent traffic statistics. Thus it would be worth to first isolate this group of cars in order to estimate their contribution to the inconsistent traffic loads in specific road segments. Preferably, this information would then be taken into account in conjunction with a database of consistent traffic statistics, preferably updated with current real time updates of traffic, to determine current and predicted traffic information that would be currently updated accordingly. The isolation process would use prediction queries that would selectively target cars that made a change to their route or deviated from schedule, according to traffic information or other predetermined possible reasons such as a response of drivers to a telematics application. The queries determine the response criteria which will include but not be limited to the following a) vehicles that are planning to pass in a certain road at a certain forward time interval according to their modified route plan or schedule, and which did not plan to do so according to a reference route a tb default route or any other route that could be reterred by the PMMS as a reference ;that may be determined according to criteria as part of a predetermined protocol), and b) vehicles that did plan to pass in this road according to the reference route, 17" and are not planning to do so according to the modified route plan or schedule at the above forward time interval.

Vehicles which are using their reference default) route will not respond to queries.

00 0 Criteria for determining whether a route is within reference conditions O io default) or not, could be provided from a common external source, which considers 7- the investigated level of possible effect on the traffic statistics. The reference Sdefault) route information may be formed either in the in-car (on board) systems, or received from external (off board) sources, and would preferably be determined by route plan and schedule. Thus, according to a predetermined protocol, a deviation in route or schedule would exclude the route from being referred to as a reference route and would determine it to be a non reference route. The protocol would preferably include threshold levels of deviation.

The in-car system will incorporate a predetermined decision procedure, described in the following.

In principle, a Differential Traffic Load Prediction (DTLP) process with respect to a Forward Time Interval related Route Segment (FTIRS refers to a time interval with respect to a route segment, usually a road segment) under investigation, could be implemented by means of two types of traffic prediction queries which would be transmitted by a mapping system to the PMMS units. The prediction queries include the prediction criteria, and are aimed at targeting groups of cars that are either expected to pass through the FTIRS under investigation and were not expected to do so, according to database information, (non expected vehicles NEV), or are not expected to pass through the FTIRS under investigation, and were expected to do so, according to the database information (expected vehicles -EV) Query A) type of a query with the aim of estimating the number of vehicles which on their reference route are not expected to pass through the investigated FTIRS, and on their non reference route are expected to pass through the investigated tFTIRS, (non expected vehicles NEV), and ;Query B) type of a query with the aim of estimating the number of vehicles which on their reference route are expected to pass through the investigated TIRS and on s their non reference route are not expected to pass through the investigated FTIRS, (expected vehicles EV).

In order to enable responses in relation to forward time intervals, it is required that the PMMS units would be equipped with the means of reference or mean to 00 0 calculate reference to segments of planned routes and estimated travel time intervals O 0 along respective route segments. Preferably, an estimated time interval will be provided with respective confidence intervals.

SVehicles which are using a non reference planned route, will enable the r response procedure according to the following decision procedure; If the received query is identified as Query A, then, according to the following differential traffic load match process result, if there is a match between FTIRS in the query and the planned non reference default) route (route in use), and there is no match between FTIRS in the query and the reference route, then enable the response procedure.

In order to alleviate the computation load in the in-car system, involved in frequent matching in response to above queries, it would be preferable to refer 3o routes to predetermined area zones, and by a preliminary predetermined screening procedure, preceding the above decision procedure, vehicles whose planned (reference and non reference) routes do not cross area zones in which the FTIRS is included, will not continue with the more detailed matching process in the above decision procedure.

A number of communication slots will be preferably allocated for responders (cars which transmit in the allocated slots) in the response procedure, separately, with respect to each Query. Each of the targeted vehicles, (responders), in which the (NI response procedure is enabled, will use a predetermined response procedure to b select a slot in which to respond. This predetermined procedure would preferably use a uniformly distributed random selection of a slot out of all the allocated slots, to transmit a signal.

A predetermined estimating procedure will be used in the non mobile system platform, to determine estimated number of responders according to the total 00 number of slots in which responses are detected in a given number of allocated slots.

The estimating procedure would preferably use a number of secondary procedures, S 10 as described in the following and illustrated in Fig 1. It is preferably aimed to obtain the estimated number of responders with an acceptable error level, however the error Slevel is a function of the ratio between the number of responders and the given number of allocated slots. The greater the number of allocated slots in proportion to the number of responders, the lower would be the error level. The error level can be defined as the maximum cumulative probability that could produce a similar result from a number of responders which is either greater or lower than the acceptable estimation interval of responders. The acceptable error level would preferably be determined according to the sensitivity of the estimation in the specific application.

Since there is a variation around the most frequent number of responding slots, (slots in which responses are detected), which depends on the number of allocated slots and the number of responders, it is desirable to assess in advance a realistic anticipated range of numbers of responders, in order to determine a minimal number of initial allocated slots for an acceptable variance. Since such realistic ranges of responders could be anticipated from statistical data, according to time and place, then a database of possible initial ranges would preferably be evolved for any particular urban entity, (preferably as probability distribution from which ranges of confidence intervals could be derived). The database of ranges would be preferably evolved taking into account conditions specific to such an entity, such as, (but not limited to), characteristic traffic conditions, characteristic infrastructure servicing traffic flow, and prevailing decision processes used by route guidance procedures.

The technique of evolving a database of ranges for initial numbers of expected responders would preferably be based on statistical and empirical methods and computer simulations. In order to determine the required initial number of allocated slots, based on the database of ranges, it is also preferably required to take into account the prevailing conditions in available radio communication spectrum, limitations imposed by the need to investigate preferred number of FTIRS in a reasonably meaningful short cycle time, and an acceptable tolerable error in the (N resulting predictions. Since the initial determined number of allocated slots might not tfoachieve the preferably acceptable error level, successive repetitive iterations in allocation of slots and re-estimation of number of responders, might be required. In order to determine the possible need for adjustment of number of allocated slots in a minimal number of iterations, an error estimating function, and an optimized adjustment function, would preferably be evolved. The error estimating function would preferably estimate the error, by confidence interval) in the resulting 00 estimated number of responders, as a function of the- ratio between the number of detected number of responding slots (responses) and number of allocated slots (preferably considering the probability distribution of responders). Based on the error Sestimating function, the required preferred number of allocated slots may have to be adjusted for a further iteration, and may also vary during a possible series of iterations. The optimized adjustment process in arriving at the preferred number of allocated slots with a minimal number of iterations would preferably use earlier results (with a non acceptable tolerable error), to predict according to statistical combination the required improvement in the error level computing Maximum Likelihood Estimates or Estimates), and to determine accordingly the preferred required number of allocated slots to be used in the subsequent iteration, in order to save further iterations. The significance in performing iterations is, in addition to the potential in reducing the error level, in checking consistency, particularly in cases where little, or no, a-priori knowledge exists about the probability distribution of responders that provide a certain number of responses. Thus, at least two iterations would preferably be allowed even though the first proportion between the number of responses and allocated slots might be satisfying, indicating on an acceptable error level.

The estimating procedure would preferably use statistical methods which could produce acceptable estimation intervals (based on interval estimation approach such as confidence and tolerance intervals with upper and lower limits). A single point that is the most frequent number of responses (responding slots) in a pre-determined number of slots for pre-determined simulated (or analytically calculated) number of responders could provide the distribution of the number of responses around this point and could determine a tolerance interval for the interval estimate. The most frequent number of responses will be referred to in the following as single point estimate for the number of responders in a predetermined number of slots. One conservative way of determining an acceptable estimation interval for decision making about the possible range of responders that respond by a certain number of responses in a predetermined number of allocated slots, is by first N determining a tolerance interval according to a respective single point estimate, tb either produced by a simulation of responses according to a certain repeated number ;Z of responders in certain number of allocated slots or by analytical calculation, then, to determine according to the response distribution of the responses an acceptable tolerance interval. Based on the acceptable tolerance interval it is enabled to determine, either by simulation or by analytical calculation, two other response distributions for the same number of allocated slots which indicate on the potential of an upper and lower number of responders to produce responses within the acceptable tolerance interval, by determining acceptable error according to cumulative-probability o1 the overlap (analogous to error type 11 in hypothesis testing, with respect to an acceptance region As a result of the single point estimates of the upper and lower distributions of responses which overlap with the tolerance interval within an acceptable error it would be enabled to determine upper and lower numbers of responders which could be used to further determine upper and lower is limits to an acceptable interval for the estimation of potential responders that might produce the same number of responses in the allocated slots. The upper and lower limits of this interval could be determined with respect to the sensitivity of the, decisions that have to be taken accordingly. Such limits could also be interpreted as determining the rejected regions of potential responders. From the point of view of the acceptable estimation interval definition, for a significantly wide range of different numbers of responses for a sufficient number of slots, consistency in terms of percentage of error would be expected around said single point estimates for a respective range of responders due to close to linear relation between said single point estimates and respective responders in that range. An alternative approach to determine estimation intervals is by producing probability distribution function (POF) of potential responders around a said single point estimate, either analytically or by simulation, from which the acceptable estimation interval could be derived e.g., according to the confidence interval of this POF. Such a PDF could be used for traffic behavior analysis according to different criteria, criteria which characterize reaction of mobile units to telematics applications, which may cause erratic traffic.

Each PDF could be derived for a certain number of allocated slots by normalizing simulated distributions of the relative frequency of a certain number of responses, determined by a said single point estimate related to a certain number of responders, which may be produced with other (lower) relative frequency by responders which have a different number from the number of responders which relates to the said point estimate. A sufficiently high range of the number of responders should be used to enable the normalization of the relative frequencies of the responses to determine (N a said PDF. For high accuracy of the relative frequencies that should be determined tb also for high number of potential responders (theoretically unlimited but practically ;limited by the application) a sufficiently high number of repetitions of response procedures should be used, to determine the relative number of the responses, for the said number of responses determined by the said single point estimate of responders (tested according to a number of allocated slots). Repeating the simulation for a sufficient range of numbers of responders to provide relative 0 frequencies of the same number of responses around relative frequency derived according to the said single point estimate would determine a distribution of the said 1o number of responses according to the (practical) range of numbers of the potential responders. According to the accumulated number of responses that produce the relative frequencies of responses (according to the said sufficiently high number of repetitions to the same number of responders) a normalization phase can be taken to produce a said PDF. The simulation could be further expanded to determine such is distributions for different numbers of allocated slots around different numbers of responders (determined by said single point estimate). Such PDFs could be used to provide confidence intervals for single estimate of responders with single allocation of slots. For estimates that would use more than a single allocation of slots it would be valuable to create joint PDFs for combinations between different numbers of slots with different numbers of responders related to the said single point estimates. Error estimating functions could further be formulated according to statistical methods and by simulations that could consider a-priori knowledge about the probability distribution of responders (Bayesian approach). The estimating process would count the number of the slots that were detected to be used by at least one responder and will use this number as an input to a predetermined estimating function based on pre stored table that includes PDFs, confidence intervals, and upper and lower limits of said acceptable estimation intervals, constructed according to simulations) which could provide required estimates as a function of number of slots detected to be used by responders in the allocated slots. The estimate would be considered as the estimation of the number of vehicles according to the query criteria. Estimating functions (tables) could be predetermined preferably by using the described method for simulation and other statistical methods known in the art. Separate estimating functions would be preferably evolved for various ranges of numbers of allocated slots. An increase in the number of allocated slots ought to shorten the acceptable estimation interval. In practice this would enable to use more efficiently the allocated communication resources. Response and detection procedures could further include a possible discrimination between number of responders in each slot. However this (N would require accurate power control on the transmitters of the responders which for tbshort burst transmissions could be more costly to be implemented CDMA). Non information signals would be preferably used by the responders. However, if information bearing signals are used by the responders capture effects also could be considered to distinguish between slots. Nevertheless short energy burst in slots could minimize time of detection and hence preferably fit to the response procedure where responders use allocated slots randomly by the responders and the detection 00 process of their transmitted signals could consider just energy detection.

The estimations that may according to one type of query selectively represent C io additional number of vehicles that were not expected (preferably according to probabilistic levels) to arrive to the FTIRS, (NEV), and according to a different type of Squery, the number of vehicles that were expected (preferably according to probabilistic levels) to arrive to the FTIRS and would not arrive to the FTIRS, (EV), would indicate on change in expected load, in the FTIRS. This could be used in conjunction with an off line database of traffic statistics to determine according to the expected traffic and the non exlected traffic (predicted differential traffic load) the weighted sum of the missing EVs and the additional NEVs with the predictable traffic load in the segment of road by using statistical methods known in the art such as convolution between PDF of the estimate of the expected load in the database and the estimated number of NEVs, would provide a PDF of the updated estimate to be used for the computation of a new expected load due to NEVs).

For this purpose it would be useful to construct respective PDF's in conjunction with the function tables that are produced to provide estimation intervals, as further described in the detailed description.

This is the basis for an improved way to predict traffic in conjunction with off line database statistics, preferably with such that are being adaptively corrected by mapping of the current traffic.

In addition to the contribution potential of such improvement to central control on traffic it would have the potential to improve, and even enable, reliable dynamic route guidance. However the way of how to use such predictions is a very important issue when considering the extensive use of car navigation systems, In which the planned routes are being independently modified according to such predictions. The following highlights a preferable method by which such predictions could enable efficient distributed DRG.

In order to explain the benefit of this approach for implementing distributed DRG it would be worth to describe traditional approaches in comparison.

~23 SIn order to overcome unpredictable traffic problems, in the future, traditional approaches are considering a system that would be almost fully controlled, in-car computers will not make the decisions for their best route but rather a Big Brother rapproach will do it by providing the recommended routes in order to maintain s predictive traffic. This approach would use a central computation method that will have to maintain the knowledge on the destination of each vehicle as well as its current position along the road. Beside the numerous computations that it would 00 require it would need a communication platform that would have to accommodate a huge volume of data that will connect the vehicles to the control center. In practice, N io roadside beacons that have two way communication capabilities are considered for this purpose. Apart from the non privacy characteristic of such a system it will have a Stremendous cost and will require computation power that probably makes the idea impractical for wide coverage implementation. This problem increases when a significant number of drivers would not obey the central route guidance, and hence it will reduce the system efficiency and could even make it unreliable. For such reasons a concept of predictive Dynamic Route Guidance based on distributed intelligence should preferably be used whereby in-car computers would be making decisions on their preferred routes. However, with such an approach the traffic would probably become even more unpredictable. To overcome this problem there would be a need to cope with unpredicted traffic in a way such as proposed above and to use periodical corrections to statistical traffic databases. To realize such an approach, predicted traffic information would have to be, periodically, estimated and then provided to the car navigation computers so that a trial and fail based process would be used to refine an equilibrium between the individual needs and the offered traffic routes. This would implement a system based on distributed intelligence in which, in addition to taking into account current traffic information, the car navigation computers would have to use a predetermined give-up process which, according to the predicted traffic information and their planned route, each car would try to identify if its planned route is going to take part in a predicted traffic congestion or traffic jam.

The identification of such situation would result from a comparison between the predicted traffic information and the planned route. If the comparison would identify predicted traffic congestion along the planned route it would automatically give up on its planned route, if it would have a more reasonable alternative route. The give up process would preferably be used according to priorities and could consider various criteria levels. For example, in a first iteration of such trial and fail cycle, cars that would have an alternative route that might increase the length of their planned route by, say 5 percent, but would not significantly affect their traveling time, would t-1 automatically change their planned route to the alternative route which a-priori had a tlower priority. A further cycle of prediction and update to the cars, probably indicating on changes in traffic predictions according to the reactions of cars to the previous give up procedure cycle, could either result in additional cars, with a higher grade of give up level alternative route with say 10% increase in length to remainder of planned route), to give up on the planned route, if previously predicted traffic congestion still predicted. Such procedures might, some times, allow cars to return to 00 an earlier, more preferable, route (reduced grade of give up level), in the case that too many cars have given up on their planned routes at a previous iteration, and o 0 accordingly traffic loads are alleviated. In addition to predetermined give up process based on parameters of increase and reduction of give up levels, random parameters Smight preferably be used in order to refine, and even to control the convergence iterative process. As a result of a sufficient number of such iterations, this process could lead to a convergence to equilibrium, with the grade of give up level and its reduction tapering off. Trade off between low and high levels of give up grades would preferably be taken into account, with the parameters of the iterative process.

When Car Navigation System (CNS) with on board DRG capability are considered as being used it would be easy to observe the benefit of such approach since periodical process of such prediction processes could help to refine the preferred route by on board DRG of the CNS units. However one of the trends in telematics is to provide off board DRG to Telematics Computers (TC) installed in cars. Such TC would be provided with a recommended route and according to in-car positioning means the TC could navigate the driver along the route. Thus to enable handling the traffic predictions in an environment that partially use TC with off board DRG and another part uses CNS units with on board DRG it would be necessary to provide enhanced capability to TC units. For example, a TC will be provided with a few alternative routes, bypass segments of routes), in order to overcome possible traffic load problems in predetermined segments investigated in the prediction process. These alternatives, would be used, according to priorities by the TC, that would be equipped with a radio interface, such as used with the CNS having on board DRG, enabling it to participate in prediction processes. Thus, by participating in the prediction processes the route plan would be refined by using a give up procedure, according to a balance between current and predicted traffic.

The predicted information would be preferably provided through a broadcast channel, RDS/TMC, to car navigation end users and off board DRG service providers as well as to traffic control centers.

Another embodiment of the implementation the differential traffic prediction 1 process deals with effects on traffic loads as a result of telematics applications, such b13 t as Local Based Services. One type of Such telematics application is position related commerce service, sometimes named as p-commerce, m-commerce or I-commerce.

With such a service application, a service user would preferably initiate a request to locate points of interest according to criteria. For example a request may ask for locations where a certain product may be found, with possible restrictions to some range of prices and possibly within a certain distance from the user's position.

00 Another application of telematics is more advertisement oriented and could be Initiated by a vendor who wishes to provide ordinary or special offers to drivers possibly for a short term. In order to enable the vendor to administer such offers efficiently it would be valuable to have a priori knowledge about the potential demand Sfor an offer. One way to get such information is to use recorded information of requests initiated by the potential buyers to assess the demand potential for a certain level(s) of price. A problem, involved with special offers, could be the lack by 1s vendors of a priori knowledge about potential buyers who might otherwise show interest in many different products, other than those, subject to a special offer.

Beside the effect of p-commerce on the traffic load there might be different ways to implement p-commerce and hence to increase the level of unpredicted traffic. For example in order to improve p-commerce applications, it would be an advantage to large stock holders and others to have a query tool that would help them to identify sufficient demand, preferably according to prices and including non solicited products, for special offers. This could create a hunting trip environment.

With such a tool, queries could be provided in a way similar to an auction process, preferably by a broadcast message to the telematics users, with respect to products with possibly one or more ranges of prices. The user, usually a driver, will have a stored list of preferences for products, in his Telematics Computer (TC) that would be matched with broadcast messages according to preferences in the list. For example, a stored product list (SPL) which may include products with ranges of prices could enable the TC to respond to a broadcast query. If such responses would provide information about the estimated number of the potential clients and possibly their position distribution it would enable the vendor to determine a time window and price for a special offer according to demand. The offer could then either target the potential clients and possibly others. Most probably this would target the responders who would contribute to the decision making. When considering a system platform with capabilities such as suggested for a traffic mapping system and telematics mobile unit with PMMS capabilities, (that uses pre assigned slots to determine position and other distributions of responders according to queries, and possibly to

Claims

1. A method of estimating a demand for an item, wherein the item is to be offered ;according to said demand, the method comprising: transmitting to mobile units information related to said item; and receiving a response generated by a mobile unit of said mobile units, wherein the response is generated according to a match process between the information Nrelated to said item as received by said mobile unit and information related to said OO item as stored in a list of items in said mobile unit. O

2. The method of claim 1, further comprising estimating the demand for said item based on a number of responses received from one or more of said mobile units.

3. The method of claim 1 or 2, wherein said response is generated according to a match between an identification of said item according to said information received by said mobile unit and an identification of said item according to the information stored in said list.

4. The method of any one of claims 1 to 3, wherein said response is generated according to one or more preferences included in the information stored in said list.

The method of any one of claims 1 to 4, wherein said list comprises one or more item identifiers associated with one or more prices, respectively.

6. The method of any one of claims 1 to 5, wherein transmitting comprises broadcasting said information related to said item to be offered.

7. The method of any one of claims 1 to 6, wherein transmitting information related to said item comprises transmitting a query including the information related to said item, wherein the query is to determine a number of potential clients for said item in a defined area.

8. The method of any one of claims 1 to 7, wherein transmitting information related to said item comprises: transmitting a query including the information related to said item, wherein the query is to determine a number of potential clients for said item in relation to their respective time of arrival.

9. The method of any one of claims 1 to 8, further comprising: transmitting to at least one of said mobile units an offer to purchase said item, wherein the offer comprises an indication of an offered purchase price of said item.

The method of claim 9, further comprising: receiving an indication from said mobile unit indicating acceptance of said offer.

11. The method of claim 10, further comprising estimating the demand for said item based on a number of received indications of acceptance of said offer. -27-

12. The method of any one of claims 10 to 11, further comprising transmitting an (Ni J3 acknowledgement of said acceptance.

13. The method of any one of claims 10 to 12, further comprising registering a user of said mobile unit to purchase said item.

14. The method of any one of claims 1 to 13, comprising controlling a number of responses from said mobile units according to one or more of the following criteria: limiting said responses to a predefined acceptable arrival range; 00 limiting said responses according to a criterion of potential arrivals of vehicles Scarrying said mobile units from one or more directions; (Ni limiting said responses according to a criterion of potential arrivals of vehicles Scarrying said mobile units through one or more route segments; (Ni limiting said responses according to deviation in predicted traffic; and limiting said responses according to whether or not a responding mobile unit is expected to change one or more of its route segments at one or more forward time intervals according to a planned route in use by said mobile unit.

The method of any one of claims 1 to 14, wherein estimating the demand comprises: estimating the demand in relation to a proportion of a number of allocated communication slots occupied by response signals from said mobile units to a number of allocated communication slots not occupied by response signals from said mobile units.

16. The method of any one of claims 1 to 15, wherein estimating the demand comprises: estimating the demand in relation to energy bursts received from said mobile units in allocated communication slots occupied by response signals from said mobile units.

17. The method of any one of claims 2 to 16, wherein estimating the demand comprises: limiting said responses according to deviation in predicted traffic in relation to one or more response criteria, including a criterion of whether or not a responding mobile unit is expected to change one or more of its route segments at one or more forward time intervals according to a planned route in use by the mobile unit.

18. The method of any one of claims 1 to 17, comprising randomly selecting by at least one of the mobile units at least one allocated communication slot for transmitting at least one response signal to a mapping system.

19. The method of any one of claims 1 to 18, comprising receiving a traffic prediction update by a mobile unit having a capability of a potential mobile mapping system.

The method of claim 19, comprising receiving said traffic prediction update by at least one unit corresponding to at least one vehicle having route guidance capability. -28- O

21. The method of any one of claims 1 to 20, wherein receiving comprises receiving at (Ni 0 least one response signal over an allocated slot selected according to a ;predetermined response procedure.

22. The method of any one of claims 1 to 21, wherein receiving comprises receiving at least one response signal over an allocated slot of a slot-oriented discrimination mapping system. C

23. The method of any one of claims 1 to 22, wherein estimating the demand comprises: 00 0 estimating the demand in relation to a sum of energy signals received from said Smobile units in allocated slots occupied by transmitted signals under a common (Ni power control of transmission.

24. The method of any one of claims 2 to 23, wherein receiving comprises receiving (Ni responses in conjunction with differentially predicting load of traffic on a route segment related to a forward time interval.

The method of any one of claims 1 to 24, wherein predicting comprises estimating the number of vehicles traveling according to a non-reference route plan in relation to at least one route segment and at least one forward time interval.

26. The method of any one of claims 1 to 25, wherein receiving comprises receiving one or more responses resulting from a mobile give-up process according to which if a mobile unit identifies predicted traffic congestion along a planned route of the vehicle, and if the mobile unit determines that the vehicle has one or more alternative routes, then the mobile unit modifies the planned route of the vehicle to an altemrnative route with a lower priority.

27. The method of claim 26, wherein the route plan of the vehicle associated with the mobile unit is modified to an alternative route having a higher priority based on one or more criteria including traffic alleviation resulting from the give-up process.

28. The method of any one of claims 1 to 27, wherein the item to be offered comprises a product to be offered.

29. The method of any one of claims 1 to 27, wherein the item to be offered comprises a service to be offered.

30. A communication system for estimating a demand for an item, wherein the item is to be offered according to said demand, the system comprising: a transmitter to transmit to mobile units information related to said item; and a receiver to receive a response generated by a mobile unit of said mobile units, wherein the response is generated according to a match process between the information related to said item as received by said mobile unit and information related to said item as stored in a list of items in said mobile unit. -29-

31. The communication system of claim 30, comprising a computing platform to estimate, based on a number of responses received from one or more of said mobile units, the tbd ;demand for said item.

32. The communication system of claim 30 or 31, wherein said response is generated according to a match between an identification of said item according to said information received by said mobile unit and an identification of said item according to the information stored in said list. 00

33. The communication system of any one of claims 30 to 32, wherein said response is generated according to one or more preferences included in the information stored in (,i S 10 said list.

34. The communication system of any one of claims 30 to 33, wherein said list comprises (,i one or more item identifiers associated with one or more prices, respectively.

The communication system of any one of claims 30 to 34, comprising a computing platform to control the number of responses of said mobile units according to one or more of the following criteria: limiting said responses to a predefined acceptable arrival range; limiting said responses according to a criterion of potential arrivals of vehicles carrying said mobile units from one or more directions; limiting said responses according to a criterion of potential arrivals of vehicles carrying said mobile units through one or more route segments; limiting said responses according to deviation in predicted traffic; and limiting said responses according to whether or not a responding mobile unit is expected to change one or more of its route segments at one or more forward time intervals according to a planned route in use by said mobile unit.

36. A method of controlling a number of responses from mobile units, wherein an item is to be offered to potential customers through said mobile units, according to one or more of the following criteria: limiting said responses to a predefined acceptable arrival range; limiting said responses according to a criterion of potential arrivals of vehicles carrying said mobile units from one or more directions; limiting said responses according to a criterion of potential arrivals of vehicles carrying said mobile units through one or more route segments; limiting said responses according to deviation in predicted traffic; and 30 limiting said responses according to whether or not a responding mobile unit is (Ni expected to change one or more of its route segments at one or more forward ;time intervals according to a planned route in use by said mobile unit.

37. The method of claim 36, wherein the item to be offered comprises a product to be offered.

38. The method of claim 36 or 37, wherein the item to be offered comprises a service to OO 0 be offered.

39. The method of any one of claims 36 to 38, comprising limiting said responses according to deviation in predicted traffic in relation to one or more response criteria, 10 including a criterion of whether or not a responding mobile unit is expected to change one or more of its route segments at one or more forward time intervals according to a planned route in use by the mobile unit.

The method of any one of claims 36 to 39, comprising receiving one or more responses resulting from a mobile give-up process according to which if a mobile unit identifies predicted traffic congestion along a planned route of the vehicle, and if the mobile unit determines that the vehicle has one or more alternative routes, then the mobile unit modifies the planned route of the vehicle to an alternative route with a lower priority.

41. A method of estimating a demand for an item, substantially as herein described with reference to any one of the embodiments of the invention illustrated in the accompanying drawings and/or examples.

42. A communication system for estimating a demand for an item, substantially as herein described with reference to any one of the embodiments of the invention illustrated in the accompanying drawings and/or examples.

43. A method of controlling a number of responses from mobile units, substantially as herein described with reference to any one of the embodiments of the invention illustrated in the accompanying drawings and/or examples.