Page 426 - From GMS to LTE
P. 426

412  From GSM to LTE-Advanced Pro and 5G

            algorithm cannot find a good value on its own. APs by some manufacturers record the
            speed at which individual stations send their frames and then use the same speed for
            subsequent packets to the clients. Beacon frames are usually sent at a speed of 1 or 2
            Mbit/s. This allows even distant devices to detect the presence of an AP. However, this
            behavior is not mandatory and some APs transmit their beacon frames at a speed of 11
            Mbit/s. This increases the overall speed of the network slightly, but has some disadvan-
            tages for distant devices.
             For the coding of the actual user data in a frame, the direct sequence spread spectrum
            (DSSS) method is used for transmission speeds of 1 and 2 Mbit/s. Instead of transferring
            the bit itself, the DSSS algorithm converts the bit into 11 chips, which are then transmit-
            ted over the air. Instead of sending a bit with the value of ‘1’ the chip sequence
            ‘0,1,0,0,1,0,0,0,1,1,1’ is transmitted. For a bit with the value of ‘0’, the sequence is
            ‘1,0,1,1,0,1,1,1,0,0,0’. These sequences are also known as Barker code. As 11 values are
            transmitted instead of only one, redundancy increases substantially. Thus, a bit can be
            received correctly even if some of the chips cannot be decoded correctly at the receiver site.
             UMTS also makes use of this technique, also known as spreading, to increase redun-
            dancy. However, there is an important difference. In a WLAN network, only a single
            station sends at a time (time division multiple access). UMTS additionally uses spread-
            ing to allow several devices to send at the same time (code division multiple access).
            This is possible in UMTS, as shown in Chapter 3, as orthogonal codes are used instead
            of fixed Barker sequences.
             Once a bit has been converted into chips, the Barker chip sequence is sent over the air
            using Differential Binary Phase Shift Keying (DBPSK) with a transmission speed of 11
            Mchips/s. The resulting bit rate is 1 Mbit/s. To transmit chips, DBPSK changes the
            phase of the signal for a ‘1’ chip by 180 degrees. For a ‘0’ chip, no phase of the carrier
            frequency remains unchanged.
             To achieve a bit rate of 2 Mbit/s, DBPSK is replaced with Differential Quadrature
            Phase Shift Keying (DQPSK) modulation. Instead of one chip per transmission step,
            two chips are transmitted. The four (quadrature) possible values (00, 01, 10 or 11) of the
            two chips are encoded into 90‐degree phase shifts of the carrier frequency per trans-
            mission step.
             To further increase the data transmission speed without increasing the necessary
            bandwidth, the 802.11b standard introduces Complementary Code Keying (CCK), also
            known as high‐rate/direct sequence spread spectrum (HR/DSSS). Instead of coding a
            single bit into an 11‐chip Barker sequence, CCK encodes the bits as follows.
             For a datarate of 11 Mbit/s, all bits of a frame are arranged into blocks of eight bits as
            shown in Figure 6.14. The first two bits of a block are then transmitted using DQPSK
            and are encoded in phase shifts of 90 degrees.
             The remaining six bits are used to generate an eight‐chip symbol. As six bits are coded
            in an eight‐bit symbol, the process adds some redundancy. The symbol is then split into
            four parts of two chips each, which are then modulated onto the carrier frequency as
            the phase changes.
             As the chip rate remains the same as for 1 and 2 Mbit/s transmissions, CCK raises
            transmission speeds to 11 Mbit/s. A disadvantage compared to lower transmission
            speeds, however, is the fact that there is less redundancy in the resulting chip stream.
             Different devices can send at different speeds depending on their reception conditions.
            Therefore, it has to be ensured that even devices that cannot receive high‐datarate frames
   421   422   423   424   425   426   427   428   429   430   431