«ETSI TR 101 987 V1.1.1 (2001-08) Technical Report Terrestial Trunked Radio (TETRA); Proposed Air Interface Enhancements for TETRA Release 2; Analysis and ...»
Observed Time Difference (OTD) Observed Time Difference is a method where the MS measures the time differences of signals received from separate base stations. To work accurately this method requires either tight synchronization between base stations or BS synchronization can be measured with special receivers located at known positions. In GSM system these receivers are called E-OTD LMUs. By combining results from the MS and the E-OTD-LMU with BS co-ordinates the location of the MS can be computed. Implementing OTD for TETRA would require major changes to the network and also quite significant software changes to the MS.
The location accuracy would depend on the accuracy with which the MS can measure the time differences (the current requirement being to ensure a maximum synchronization error of ¼ symbol - see EN 300-392-2, clause 7.6 ).
GPS GPS methods are based on positioning a subscriber with the additional system known as the Global Positioning System.
Stand-alone GPS Those who have used traditional Stand-alone GPS receivers may have noticed the weak points of it. When initializing the equipment, the user is asked to input an estimate of their location and time information. Typically this information is inputted as: "I am in Belgium and the time is quarter past eleven (checked from wrist watch)." The procedure starts with checking from the almanac the possible satellites that could be available in Line Of Sight. When satellites are found the receiver synchronizes itself to the direct sequence spread spectrum signal sent by the satellites. The receiver has to be able to decode the coarse acquisition information from the L1 carrier in order to have the first fix of location.
Since a low cost GPS receiver does not have a very accurate clock, the initial frequency error between the local oscillator and the satellite signals can be relatively large. A GPS receiver must step its time base systematically and attempt blind synchronization, before the local clock is tuned to the correct value. In addition, the receiver must listen to the GPS signal for a while before it can decode all needed information. Thus, a relatively long time - up to two minutes
- is needed in a traditional GPS receiver after power on until the first location fix. If there is no prior information about the coarse position, time and date needed by the GPS receiver to compute which satellites to search, the start-up can take hours.
This lengthy process may not be feasible for a handheld MS where power consumption and operating times are critical.
If provision of location is required with an emergency call, the process would be expected to be rapid. To overcome these deficiencies an Assisted-GPS has been developed. In the market there exists at least two type of assisted methods, Network Assisted - Mobile Based and Mobile Assisted - Network Based GPS.
Network Assisted - Mobile Based GPS In mobile-based GPS, the receivers in fixed positions (at Base Station sites) measure GPS and other information all the time. At appropriate and predefined intervals the network broadcasts point-to-point messages to MSs. This sent information is referred as assistance data. Assistance data includes parameters such as visible satellites at the moment, the ionospheric condition information, C/A code information, bad satellite IDs, etc.
The benefits of mobile-based GPS are related to response time and also to the required signal level. Since the receiver already has all the required information related to satellites used in location measurements, the receiver can start to calculate the position without the lengthy synchronization process. In addition, with assistance data, the GPS receiver obtains a significant gain benefit that can be used in areas where propagation conditions are weak, e.g. urban areas and inside buildings. The gain benefit is received due the fact that the S/N required for measuring the time estimates is far less than the S/N required for decoding the data transmitted by the satellite.
Mobile Assisted - Network Based GPS
The mobile-based GPS method does not significantly affect the complexity of the GPS receiver. In network based GPS, part of the functions of the GPS receiver are moved to the cellular network. The basic solution contains RF receivers, code generators and correlators to measure satellite timing. The cellular network broadcasts assistance data, which contains a list of satellites, their expected carrier Doppler shifts and code search phases. The timing measurement results are sent to the location server on the network, which performs the location calculation.
The drawback of the network-based method when compared to network assisted one is the limited capacity of the system. In a case where the location has to be updated frequently the signalling load becomes significant. The greatest advantage of the network-based method over the mobile-based method has been the simplicity of radio parts that have to be implemented into an MS. However the rapid development and the achievements on the integration level of GPS circuit technology have shifted the advantage to the mobile-based method. The savings, both in the required PCB area and in costs, are not relevant anymore.
184.108.40.206 Summary In the telecommunication standardization arena there are numerous different methods for location that have been dealt with. When making a decision on the method that should be accommodated in TETRA, following issues must be
• Accuracy requirement
• Implementation penalty of the Network
• Implementation penalty of the MS
Cell ID would be the simplest method to implement but its accuracy obviously depends on the size of the cell in which the MS is located. However, in urban areas where a greater accuracy may be required the cell size is likely to be smaller and hence implicitly provides the greater required accuracy.
For timing methods that depend upon tight synchronization between the MS and BS, the symbol duration is significant as the permissible synchronization error (which directly contributes to the RTT estimation error) is proportional to the symbol length. Tetra's relatively long symbol length, compared to the channel bandwidth, can cause a larger RTT error than GSM, leading to poorer location accuracy. Methods that rely on measuring time differences of observed transmissions are generally limited to the measuring resolution of the MS or BS element. In many cases the final accuracy (e.g. 1-2 km) of methods that are based on time information may not be significantly better than that of Cell ID.
The Angle of Arrival measurement could give relatively accurate results, but only in good conditions. In environments where reflections occur, the accuracy also suffers and the AoA method requires major changes to the network side such as co-ordination between base stations and smart antennas that would add significant costs to operators.
One of the big weakness of timing-based triangulation techniques is that for reasonable accuracy they rely on the MS being able to receive, or be received by, at least three base stations. Unless the network is specifically designed with additional base stations the probability of this being achieved will depend upon the current location of the MS as well as the network coverage planning criteria.
GPS based methods can provide the best location accuracy (10-50 m) but perform weakly in urban environment and in buildings. The lengthy start-up process for stand-alone GPS may be mitigated by implementation of an assisted method - of which the network assisted - mobile-based method seems to be most attractive.
5.5.3 Extended Range Capability 220.127.116.11 General TETRA Release 1 has a limitation in range due to timing issues. For trunked mode the range is limited to about 58 km and for direct mode the limitation is even to be less.
In some situations there is an operational need for an extended range. Requirement scenarios include aeronautical and maritime use, "linear cells" (e.g. pipelines, railways) and large rural cells (large low-traffic areas).
For land-based MSs a modified Hata propagation model should be used - see ERC report 68 "Monte Carlo Radio Simulation Methodology" (to be found at http://www.ero.dk/).
18.104.22.168 Aeronautical A typical situation for a longer range is aeronautical use over land or sea. In this situation the aeronautical mobile wants to have contact with other users on land.
Propagation studies show that at 400 MHz a range of 200 km is feasible (based on free space path loss which applies instead of the modified Hata model). Such a range would, in fact, fulfil an operational need to have communications over the whole North Sea.
To minimize the number of frequencies needed for aeronautical use there must be efficient use of the radio carriers. It is expected that up to 3 independent user groups could be active in the same area in a long-range situation.
The speed during flights is typical 200 to 400 km/h and the maximum speed of the commonly used aircraft is 500 km/h.
In normal long-range situations (with only aircraft involved) all TMO facilities are required. A terminal should be able to roam between long-range cells and normal range cells. The users should not notice a difference between long range and normal range cells.
The free space range (km) is given by (approximately) 3,5√4 where h = height (m) of MS antenna above ground.
The values in table 17 lead to achievable distances of:
MS height = 1 000 feet (300 m) 60,6 km MS height = 3 000 feet (900 m) 105 km MS height = 10 000 feet (3 000 m) 191,7 km 22.214.171.124 Linear cells Coverage along a road or pipeline would also benefit from an extended range capability. In such a scenario, minimal traffic would be generated off the linear route.
Linear cells permit the use of highly directional antennas, providing an extra 10 db of gain above normal TETRA.
In order to calculate the maximum achievable distance for a BS-MS link, the assumptions in table 19 were considered.
126.96.36.199 Large rural cells (Rural Telephony / Telemetry) Another scenario for extended range would be for communication (primarily duplex) to fixed outstations and mobile stations in low density environments.
188.8.131.52 Technical Means of Achieving Extended Range The maximum radius of TETRA base station cells is a function of the guard band between the TDMA timeslots. The maximum transmission radius is defined such that each MS transmission arrives in its allocated slot at the BS. At the BS, a transmission from a remote MS in slot 1 can collide with a transmission from a local MS in slot 2 if the round trip propagation delay is large enough (i.e. the MS is remote enough).
The current TETRA standards provide a total of 14 bits as a guard band on uplink transmissions to allow the BS to "train" to the incoming MS burst. As each bit has a duration of 27,78 µs the guard band is of 388,92 µs which in free space equates to a maximum cell radius of 58,34 km.
Range in TDMA systems in general is defined by the width of the guard band. To retain 4-slot TDMA for TETRA the guard band can only encroach for normal uplink bursts into the tail bits of the logical channels. There would appear to
be two main options to extend range:
• Only allowing use of every other timeslot, effectively extending the guard band by an additional timeslot, 255 symbols. This would theoretically extend the range, in terms of timing, by over 2 000 km, far longer than any practical propagation situation. The main advantage to this approach is that the standardization effort is anticipated to be simpler, given that the TETRA slot structure is retained. However the disadvantage of this approach is the spectrum inefficiency, half the uplink channel capacity is unused.
ETSI 39 ETSI TR 101 987 V1.1.1 (2001-08)
• Use timing advance features as is used by the GSM family of standards. The introduction of timing advance to TETRA would be a radical departure from the current slot structure, The big disadvantage of this solution is that it would require significant standardization and development effort on the air interface, impacting on terminals and base stations. The advantage of this technique is that the spectrum efficiency is retained.
It is felt that handover could be improved by introducing the following options to the air interface standard (EN 300 392-2 ).
1) Define the RSS measurement methodology in the MS. This definition will impact on other areas of the present document, notably MS open loop power control and MS location techniques.
2) Make recommendations (in an annexe) for the values of existing broadcast parameters for different cell types.
3) Add the colour code to the neighbour cell information element, clause 18.5.17.
4) Make the LA identifier mandatory in the U-Location Update Demand, clause 184.108.40.206.
5) Standardize the meaning of (at least "high") the Cell Service Level element of clause 18.5.5.
6) Change the closed loop power control descriptor (table 342) to include "near" uplink failure.
7) Change the closed loop power control descriptor (table 342) to include "near" maximum path delay.
8) Change the modelling in the downlink measurement to include a "near" downlink failure in clause 220.127.116.11.
9) Introduce a new PDU to indicate the uplink quality from the SwMI to the MS.
10)Introduce a new broadcast element to indicate the MS is to use external handover algorithms.
11)Introduce a new PDU to poll an MS for its downlink quality measurements.
12)Introduce a new PDU for an MS to report its downlink quality measurements as a response to the poll.
13 Introduce a new PDU to inform a MS to move to another cell.
14)Introduce a new PDU for a MS to respond to the instruction to move cell.
15)Introduce procedures for the use of the new PDUs in options 9 - 14. In particular such procedures must cater for half duplex calls (individual or group) and discontinuous transmission. In these cases the MS may not have transmit permission or is not transmitting even though it is allowed to.
Hierarchical Cell Structures
The following options were identified as solutions to enable the implementation of Hierarchical Cell Structures:
1) Define the "Cell Priority" identifier. This will be used to label Microcells, and get them distinguished from Macrocells. The values of the TEMPORARY_OFFSET and PENALTY_TIME parameters will be fixed in the MS. This is not a flexible solution but a quick win that enables the implementation of Hierarchical Cell Structures.
ETSI 40 ETSI TR 101 987 V1.1.1 (2001-08)
2) Add the TEMPORARY_OFFSET and PENALTY_TIME parameters to the broadcast message. This will add more flexibility to the use of Hierarchical Cell Structures. Different cells can be treated differently according to their size, their environment (urban, suburban or rural) and their level in the hierarchy (macro, micro or pico).
Adding these parameters to the broadcast message requires significant change to the standard and therefore should be done in phase 2.
Frequency Hopping Some work needs to be done to investigate the feasibility of implementing Frequency Hopping in TETRA. Two main
areas were identified, these are:
1) Investigate the complexity of the Linearization problem. This problem is a major challenge to the implementation of Frequency Hopping in TETRA and it needs to be solved before Frequency Hopping is given the go-ahead in the standards. A significant amount of work is needed to establish whether Frequency Hopping is viable or not.
2) Evaluate the benefits of FH and compare with the cost of its implementation and standardization. Clause 5.2.3 demonstrated some potential benefits of Frequency Hopping, but these benefits were not quantified. The extent of these benefits will need to be established to determine the feasibility of Frequency Hopping.