Patent Application
20080076435
Referring to
In operation, the mobile unit 52 transmits wireless signals 54 to the base station 60. More particularly, an antenna 57 receives the wireless signals 54 from the mobile unit 52 and transmits the signals to the base station 60 using a feed-line 59. The base station 60 processes the wireless signals from mobile unit 52 and sends the processed wireless signals 56 to the hub station 62. The base station 60 and the hub station 62 communicate using a wireless link 70 (as described below). After receiving wireless signals from the base station 60, the hub station 62 routes the processed wireless signals to the base station controller 68 using a wired communication link 64 (which may be, e.g., the Ethernet or dedicated T-1 lines or which may be a wireless link such as a microwave relay). The base station controller 68 routes the processed signals to a mobile switching center 76 which routes the communication to other subscribers on the same network or other telephones via the public switched telephone network 78. Signals can also be sent in the other direction from the public switched telephone network 78 to the mobile unit 52 using the base station controller 68, hub station 62, and base station 60.
The process for transporting signals in either direction between the base station 60 (which receives the signal from the mobile unit 52) and the base station controller 68 is referred to as “backhaul.” In system 50, the backhaul link 74 includes the wireless link 70 between the base station 60 and the hub station 62 and the wired link 72 between the hub station 62 and the base station controller 68.
In order to reduce the cost of installing, configuring, and/or maintaining a system for cellular backhaul, the base station 60 communicates wirelessly with the base station controller 68 through the hub station 62 rather than being directly connected to the base station controller 68. It is believed such a configuration can reduce the cost of cellular backhaul because the wireless base station 60 provides a method for the mobile unit 52 to communicate with the core of the network without requiring a wireline (e.g., a T1 line) or directional wireless link (e.g., a microwave relay) to be connected to each base station that receives wireless communications from the mobile unit 52.
For example, systems which do not utilize such a wireless backhaul link between a base station 60 and a hub station 62 to relay information often have a T1 or microwave link directly from the base station that receives the wireless signal to the base station controller (e.g., as shown in
By replacing the link 30 in
The base station 60 includes a transmitter 84 and a receiver 86 for communicating with the mobile unit 52 and for communicating with hub station 62. The base station 60 also includes a signal processor 100 for processing signals sent between the mobile unit 52 and hub station 62. The base station 60 can use different protocols for communicating with the mobile unit 52 and hub station 62 over wireless links 66 and 70, respectively, but use the same transmitter and receiver, antenna 57, and feed-line 59 for each of those links. This saves substantial hardware cost, since one transceiver can be used where two would ordinarily be required, and substantial operational costs, since the same antennas and feedlines may be used, eliminating incremental tower lease costs (e.g., the incremental tower lease costs associated with microwave relays).
While the same communication standard could be used to communicate with the mobile and the base station (e.g., as disclosed in U.S. patent application Ser. No. 10/256,720 filed on Sep. 27, 2002) it is believed that using different communication standards, a more efficient spectrum utilization and/or lower deployment and operational costs are realized. This is because of the differences in the requirements for the communication link between the mobiles and base stations on the one hand and the base station and hub station on the other.
Since the location of the mobile unit 52 relative to the base station 60 varies as the user of the mobile unit 52 moves, a standard wireless protocol for communication with a mobile device typically includes many signal processing techniques to mitigate the variation in the signal caused by movement of the mobile. One effect of such variation is known as a rapid fade. Measures taken in a communication standard (and the device that implements the standard) to mitigate the impact of rapid fades might include incorporation of a diversity receive path, an adaptive equalizer and/or aggressive error correction coding. These measures can add cost to a product, reduce the product's data throughput, increase the latency of a transmission, reduce its battery life, increase its power consumption and/or even add size to a product.
In contrast, because the position of the hub station 62 relative to the base station 60 is fixed, a different communication protocol can be used for sending signals between the hub station 62 and the remote station 60 than is used for sending signals between the base station 60 and the mobile unit 52. Since the hub station 62 to base station 60 link does not experience the negative effects of mobility such as rapid fades, this protocol need not employ as aggressive signal processing methods in order to maintain a communication link. These methods may be used to extend the range of operation of the system, increase its throughput or increase the reliability of the link in the face of external sources of noise or disruption of the transmitted signal. On the other hand, if these performance enhancements are not required, the static nature of the base station 62 to hub station 60 link may be used to reduce the signal processing requirements and associated costs of the link.
In some embodiments, it is desirable to use different communication protocols for wireless links 66 and 70 because the communication requirements for these links are different. For example, the communication protocol used to communicate between the base station 60 and hub station 62 (e.g., wireless link 70) does not need to account for location varying performance in the link 70 as would be needed for communication between the mobile unit 52 and the base station 60 (e.g., wireless link 66). In addition, other factors associated with mobility such as the use of specialized signaling information intended to identify and authenticate the mobile unit 52 when it enters the coverage area of a particular base station is not necessary in a wireless link between fixed locations (e.g., such as in link 70).
As a result, if the communication link 70 between two fixed base stations (e.g., the base station 60 and the hub station 62) uses a communication standard meant for communication mobile devices, the communication will be sub-optimal with respect to spectral efficiency. Thus, when communicating with the hub station 62, the base station 60 uses a waveform that is different from the waveform used to communicate with mobile unit 52. This allows the use of a more efficient communication protocol for handling the wireless backhaul link 70 between the base station 60 and the hub station 62.
In some embodiments, the communication protocol used for wireless link 70 is a custom developed protocol. The protocol uses 100 kHz bandwidth for each half duplex channel (uplink and downlink), orthogonal frequency division multiplexing, trellis coding with 4 dB of coding gain and achieves raw data rates of approximately 300 kbps. In addition, upper layers of the protocol perform MAC address translation, Ethernet packet compression and routing. The protocol also employs rate adaptation to overcome jitter effects by buffering data and transmitting such data at scheduled intervals. This step is taken in order to ensure interfaces with other systems that require predictable information arrival times can interoperate with a general purpose processing environment where execution times are not managed in a deterministic fashion.
After receiving the wireless signal from the mobile unit 52, the base station 60 processes the wireless signal according to the communication standard used by the mobile device (124). In some embodiments, the use of software based radios (for example, software radios such as those described in U.S. patent application Ser. Nos. 10/716,180, 11/071,818, 11/148,953, and 11/148,949, the contents of which are hereby incorporated by reference) can allow at least a portion of the functionality typically performed by a base station controller such as power control and/or timing advance to be performed by the base station 12. It can be beneficial to move such functionality to the base station 12 because it can reduce the backhaul bandwidth required by, for example, routing traffic that is local directly to its destination rather than employing backhaul resources to carry the traffic to the switch location and back to the serving cell.
The base station 60 modulates the signal using the protocol for communication between the base station 60 and the hub station 62 (126) and transmits the modulated signal using transmitter 84 (128).
Hub station 62 receives the wireless signal from the base station 60 using a receiver 104 (130). After receiving the wireless signal, hub station 62 de-modulates the signal (132) and transfers the signal to the base station controller 68 using a T-1 line 64 or other link (134).
In some embodiments, due to the link quality in the transmission of a signal over a wireless link 70 between the base station 60 and the hub station 62, various types of application level quality of service (QOS) and failure recovery can be desirable. In many real-time systems, TCP-style re-transmission is not appropriate, since the data may be too old by the time it is re-transmitted. Other approaches involve embedding error correction into the data stream so that lost packets can be reconstructed, and/or rules for dropping or repeating packets in the event of a loss.
One important parameter for a wireless communication is keeping the call alive (e.g., ensuring the transmission and receipt of signaling and control data used to maintain the call). In cellular systems, callers are accustomed to occasional drop outs or degradation in voice quality, but a dropped call can be a more significant problem.
In general, wireless communication protocols such as CDMA, TDMA, GSM, and iDEN are configured to expect a high bandwidth and low latency connection such as a T1 line, from the base station to the base station controller (e.g., as shown in
In some embodiments, a retransmission protocol is used to increase the reliability of the wireless link 70 and reduce the frequency with which the wireless link 70 causes a loss of connection to the wireless call (e.g., reducing how frequently a cellular call is ‘dropped’ by the network). The retransmission protocol is based on an acknowledgement scheme in which the hub station 62 informs the base station 60 when a packet has been successfully received.
In order to implement the retransmission scheme, the wireless signals can be categorized into different classes which are used to determine whether or not to re-transmit a packet. The wireless traffic is categorized as signaling/control data or payload data. The signaling/control data is data used to maintain the call. Examples of such data include handover, power control and timing advance. If the signaling/control data is not received by the hub station 62 and retransmitted to the base station controller, the wireless link will fail and the mobile unit 52 will experience a dropped call. In contrast, payload data is data such as the voice data in a wireless call. If a portion of the payload data is not received successfully, the user of the mobile unit 52 may experience some noise in the call but the link typically will not fail. Since the signaling/control data is needed to maintain the call, the signaling/control data can be assigned a higher priority for retransmission than the payload data.
As shown in
In addition to the re-transmission protocol described above, various other mechanisms can be used to ensure the latency and quality of the signal transmitted from the mobile unit 52 to base station controller 68 over the wireless links 66 and 70 is maintained. Since the wireless link 70 has higher latency and increased error rate compared to a T1 link, it can be beneficial to use various techniques to ensure that the quality-of-service (QoS) is maintained such that there is not an interruption in the voice service for the cellular customer. For example, the protocol implements a selective repeat procedure, which allows for a single retransmission of certain packets, in the event certain packets are not delivered error-free. An error-free delivery determination is made by reference to CRC (cyclic redundancy check) in the event of a packet that has arrived or with reference to timing requirements or packet sequence numbers in the event of a packet that fails to arrive.
As shown in
Such a hub-and-spokes arrangement can be beneficial because the overall area covered by the wireless system 51 can be increased without requiring as many wired connections. Since fewer wire-based communication links are needed, the cost of operating a hub-and-spokes based network 51 utilizing a wireless backhaul link 70 can be lower than operating multiple base station units each connected directly to the base station controller 68. Because the hub station 62 may be shared by many base stations 60 for backhaul of wireless signals, the cost of the link 64 from the hub station 62 to the base station controller 68 may be spread over a number of base stations 60.
For example, as shown in
In some embodiments, as shown in
Routing the backhauled information to a particular base station based on the frequency of transmission can reduce the latency caused by backhaul transmission compared to the use of a higher layer routing protocol. In general, a higher layer routing protocol would require, for example, demodulation of the signal to determine the address(es) to which individual packets are to be routed. This demodulation would result in a greater latency in comparison to routing the signal based on the frequency of the communication.
Because the waveforms, transmitters, and receivers employed to perform backhaul are software applications, it is possible to reallocate wireless resources, including backhaul resources, dynamically. Thus, it is possible to reallocate some or all communications channels and backhaul channels from an idle base station to another base station with additional capacity needs. For example, if no mobile stations were attached to base station 210, frequency f1 can be redirected to base station 212 to temporarily increase the capacity of base station 212.
In addition to frequency of operation, other examples of physical layer information that could be used to route the backhauled signals include: timeslot of transmission (on a shared channel), and/or orthogonal code in the case of a CDMA based backhaul system. Signals transmitted by the hub station 220 may be repeated at a base station in order for them to reach a further base station that is the addressee of the backhauled signal.
In some embodiments, as shown in
As shown in
The system can also manage jitter introduced into the system as a result of the backhaul transmission by buffering. For example, in some embodiments, the system can include a jitter buffer at one or both ends of the backhaul link to compensate for jitter in the shared network. In general, signal processing systems include some jitter which is a random variation in the time required to complete any particular task. At the lowest levels of the system, the jitter is due to hardware effects, such as the relative time at which two chips request access to a shared bus. At higher levels, the jitter comes from variable and unpredictable network performance. The jitter buffers can ensure that the system will continue to process signals and present them to the system users in accordance with the relevant communications protocol even when significant jitter exists in the network. The buffering employed in the protocol adapts based on performance of the link in question. Within limits, it will employ longer buffers if there is no data available for transmission out of the buffer at the scheduled time for transmission. On the other hand, if the system is performing well (no missed transmissions), the protocol will shrink the buffer in order to decrease end to end latency. The protocol may also employ methods for assigning priority to, and scheduling accordingly, the transmission of data out of its buffer in order to optimize overall system performance by minimizing the likelihood of collisions between packets transmitted simultaneously by multiple stations or by assigning higher priorities to certain packets (e.g., control packets) than other packets.
Other implementations are within the scope of the following
