United States Patent [19]
`Stjernholm
`
`[54] METHOD FOR ESTIMATING TRAFFIC
`DENSITY IN MOBILE TELEPHONE
`NETWORKS
`Inventor: Paul Stjernholm, Stockholm, Sweden
`[75]
`[73] Assignee: Televerket, Farsta, Sweden
`[21] Appl. No.:
`185,973
`[22] PCT Filed:
`Jun.24, 1993
`[86] PCT No.:
`PCT/SE93/00575
`§ 371 Date:
`Feb. 10, 1994
`§ 102(e) Date:
`Feb. 10, 1994
`[87] PCT Pub. No.: W094/05098
`PCT Pub. Date: Mar. 3, 1994
`Foreign Application Priority Data
`[30]
`Aug. 18, 1992 [SE]
`Sweden ................................ 9202369
`
`[51]
`
`Int. Cl.6 ..................... H04M 3/22; H04M 11/00;
`H04M 15/00; H04Q 9/00
`[52] U.S. Cl •...................................... 379/112; 379/34;
`379/58; 379/59; 379/111; 379/113; 455/33.1;
`455/33.4; 455/54.l; 455/67.1
`[58] Field of Search ....................... 379/32, 34, 59, 58,
`379/111, 112, 113; 455/33.1, 33.4, 54.1, 56.1,
`67.1
`
`[56]
`
`References Cited
`U.S. PATENT DOCUMENTS
`4,670,899 6/1987 Brody ................................. 455/56.l
`4,701,934 10/1987 Jasper .
`4,706,286 10/1987 Sturza .
`5,043,736 8/1991 Darnell ............................... 342/357
`5,056,106 8/1991 Wang et al ..
`5,095,500 3/1992 Tayloe et al .......................... 379/32
`5,157,709 10/1992 Ohteru .................................. 379/58
`5,175,867 12/1992 Wejke et al ........................ 455/67.1
`5,201,061 4/1993 Goldberg ........................... 455/67.1
`5,263, 177 11/1993 Schieve .............................. 455/56.1
`5,268,694 12/1993 Jan et al. ............................ 455/33.4
`
`I lllll llllllll Ill lllll lllll lllll lllll lllll lllll lllll lllll llllll Ill lllll llll
`US005418843A
`5,418,843
`[11] Patent Number:
`[45] Date of Patent: May 23, 1995
`
`5,289,183 2/1994 Hassett ............................... 455/54.1
`5,293,642 3/1994 Lo ...................................... 455/67.1
`
`FOREIGN PATENT DOCUMENTS
`0389972 3/1990 European Pat. Off ..
`0458768 4/1991 European Pat. Off ..
`2-26420 1/1990 Japan .
`2234142 5/1990 United Kingdom .
`W090/10342 7/1990 WIPO .
`
`OTHER PUBLICATIONS
`IEEE Transactions on Communications, vol. Com-32,
`No. 7, Jul. 1984, "Performance of Portable Telephone
`Using Spread Spectrum".
`Primary Examiner-Stephen Chin
`Assistant Examiner-Vijay Shankar
`Attorney, Agent, or Firm-Obion, Spivak, McClelland,
`Maier, & Neustadt
`ABSTRACT
`[57]
`A method is employed for estimating traffic density in a
`mobile telephone network. A specialized measuring
`mobile unit which is capable of determining its own
`position independently of the system is allowed to pass
`through relevant traffic routes during which time the
`measuring mobile station transmits a radius signal as
`well as its position to the various fixed measuring sta(cid:173)
`tions. The signal strength received from the measuring
`mobile station is recorded synchronously with the aid of
`a number of suitably cited measuring base stations. The
`measuring values from the measuring stations are com(cid:173)
`piled with the respective registered positions so that an
`unambiguous relationship between the signal strength
`values and the position of a mobile station passing
`through the entire system is obtained. An actual traffic
`is then measured with the aid of the measuring station
`and the measurement results thereof are related to the
`geographic positions so that a topographic distribution
`of the traffic density is obtained.
`
`6 Claims, 1 Drawing Sheet
`
`BS
`
`BS
`
`BS
`
`Apple Inc. Exhibit 1011 Page 1
`
`

`
`U.S. Patent
`
`May 23, 1995
`
`5,418,843
`
`Apple Inc. Exhibit 1011 Page 2
`
`

`
`1
`
`5,418,843
`
`METHOD FOR ESTIMATING TRAFFIC DENSITY
`IN MOBILE TELEPHONE NETWORKS
`
`FIELD OF THE INVENTION
`The present invention relates to a method for estimat(cid:173)
`ing traffic density in mobile telephone networks, mobile
`radio networks and the like. The method is based on
`synchronised signal strength measurements in the
`uplink and estimation of position, for example with the
`aid of a neural network.
`
`2
`of the positions. The more measuring stations there are,
`the greater is the probability that each position provides
`a unique set of signal strength values. Conversely, a
`certain set of measurement values corresponds to one
`S and only one position. It is this fact which is utilized in
`the present invention.
`Thus, it is necessary to establish a set of signal
`strength values for each possible position of the mobile
`station. By allowing a number of measuring stations to
`10 measure signal strength and signal identity from a mo(cid:173)
`bile station synchronously, a set of signal strength val(cid:173)
`ues is obtained at each measuring instant which corre(cid:173)
`sponds to the current position of the mobile station at
`the time. A special measuring mobile station is used
`which is allowed to drive through all relevant traffic
`routes so that the measuring result becomes sufficiently
`comprehensive and extensive. The measuring mobile
`station regularly registers its position. The measuring
`stations obtain a reference between measurement result
`and current position. It is important that the measuring
`stations are selected in such a manner that redundancy
`in the measurement results is avoided so that an unam(cid:173)
`biguous relationship between measurement result and
`position can be established. If the existing measuring
`station placements do not meet these requirements,
`further measuring stations should consequently be set
`out.
`The measurement result must then be compiled so
`that it will then be possible to estimate the position of
`the individual mobile stations. This can be done, for
`example, by using the coverage measurements which
`have been made and allowing them to train a neural
`network to translate combinations of signal strengths
`into position.
`After that, synchronised signal strength measure(cid:173)
`ments are carried out in the uplink of the actual traffic.
`The same measuring stations must be used as were used
`for the signal strength measurement. Current channels
`are measured synchronously for a relatively long time.
`Measuring stations measure the signal strength from a
`mobile station synchronously in order to generate sets
`of measurement values. Signal strength, frequency and
`signal identity are registered in order to obtain a statisti(cid:173)
`cal basis. Each combination of measurement values
`generates a position with the aid of the neural network
`and, taken together, information is obtained on how the
`traffic is distributed within the measured area.
`Since the frequency and signal identity plan is known,
`the measured measurement results can be related to a
`respective measuring station. The results are obtained in
`the form of a number of measurement results per posi(cid:173)
`tion and measuring station. By normalising against the
`SS total number of measurement results for a certain mea(cid:173)
`suring station, the traffic density of the measuring sta(cid:173)
`tion is obtained.
`The traffic density per measuring station can also be
`weighted with traffic management. This can be mea(cid:173)
`sured or retrieved from statistics from the exchange. By
`adding the weighted traffic densities of all the measur-
`ing stations, a general view is obtained of how the traf(cid:173)
`fic is distributed in the entire network. If this distribu(cid:173)
`tion is normalised, the traffic density for the entire net(cid:173)
`work is obtained.
`By simply making use of the measurement results
`and/or weights for certain time intervals, busy hour,
`etc., an understanding can be obtained of how the traffic
`
`PRIOR ART
`A natural desire for an operator of a mobile telephone
`network is to obtain information on how the traffic is lS
`distributed geographically in the network. The informa(cid:173)
`tion can be used for being able to plan the network with
`respect to capacity and with respect to quality.
`It is know per se to measure signal strengths at base
`stations during a call in progress with a mobile station. 20
`See for example EP-A-431,956. However, it has been
`found to be difficult to obtain the actual geographical or
`topographic distribution of the traffic since it is difficult
`to determine the positions of the individual mobile sta-
`tions.
`
`2S
`
`SUMMARY OF THE INVENTION
`The present invention relates to a method for estimat(cid:173)
`ing traffic density in mobile telephone networks, which
`method solves the above-mentioned problems. Accord- 30
`ing to the invention, a measuring mobile station is al(cid:173)
`lowed to pass through relevant traffic routes while it is
`sending a radio signal and at the same time recording its
`position. The signal which is received from the measur(cid:173)
`ing mobile station is measured in synchronism with the 3S
`aid of a number of suitably sited measuring stations. The
`measurement values from the measuring stations are
`then compiled with the respective registered positions
`of the measuring mobile station so that an unambiguous
`relationship between signal strength values and position 40
`is obtained.
`After that, the actual traffic in the mobile telephone
`network is measured with the aid of the measuring
`stations and the consequent measurement results can
`then be related to geographic positions with the aid of 4S
`the compilation so that a topographic distribution of the
`traffic density is obtained. Other embodiments of the
`invention are specified in greater detail in the subse(cid:173)
`quent patent claims.
`
`SO
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`The invention will now be described in detail with
`reference to the attached drawing wherein the only
`FIGURE diagrammatically shows a mobile telephone
`system.
`
`DETAILED DESCRIPTION OF PREFERRED
`EMBODIMENTS
`The FIGURE diagrammatically shows a measuring
`system which consists of a number of measuring stations 60
`BS and mobile stations MS, only one of which is drawn.
`Nothing prevents measuring stations from being placed
`at existing base station locations and utilizing the exist(cid:173)
`ing antenna equipment. It will be seen that a signal from
`the mobile station MS gives rise to a set of signal 6S
`strengths at the measuring stations. The ratio between
`the signal strengths depends on the environment in the
`area and probably gives rise to a unique data set at each
`
`Apple Inc. Exhibit 1011 Page 3
`
`

`
`5,418,843
`
`4
`measuring the actual traffic in the mobile telephone
`network with the aid of the measuring stations, and
`relating the measurement resulting from the measure(cid:173)
`ment of the actual traffic with geographic positions
`so that a topographic distribution of the traffic
`density is obtained.
`2. A method according to claim 1, wherein said base
`stations are utilised as measuring stations.
`3. A method according to claim 1or2, characterised
`in that the measurement of the actual traffic comprises a
`number of measuring stations synchronously measuring
`the signal from a mobile station which has traffic with
`one of these measuring stations.
`4. A method according to claim 3, characterised in
`that the measurement of the actual traffic is carried out
`by all channels of affected measuring stations being
`successively passed through.
`5. A method according to claim 3, characterised in
`that the measurement values from the measuring sta(cid:173)
`tions are compiled by training a neural network.
`6. A method according to claim 3, characterised in
`that signal strength, frequency and signal identity are
`measured when measuring the actual traffic.
`* * * * *
`
`5
`
`3
`is distributed at the time and how the traffic density
`varies with time.
`The invention is only limited by the patent claims
`below.
`I claim:
`1. A method for estimating traffic density in mobile
`telephone networks, characterised by the steps:
`allowing at least one measuring mobile station capa(cid:173)
`ble of determining the location of said measuring
`mobile station geographically, to pass through a 10
`relevant traffic routes,
`determining in said measuring mobile station the posi(cid:173)
`tion of said measuring mobile station;
`allowing the measuring mobile station to transmit a
`radio signal and its determined position during the 15
`pass through the relevant traffic routes,
`measuring signal strength received from the measur(cid:173)
`ing mobile station synchronously with the aid of a
`number of suitably sited measuring stations,
`compiling the measurement values from the measur- 20
`ing stations with the respective registered position
`of the measuring mobile station so that an unambig(cid:173)
`uous relationship between the signal strength value
`and position is obtained,
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`Apple Inc. Exhibit 1011 Page 4