Скачать книгу

images

      Special cases.

       – Noiseless channel: X and Y symbols are linked, so:I(X, Y)=H(X)=H(Y)

       – Channel with maximum power noise: X and Y symbols are independent, therefore:I(X, Y) = 0

      Claude Shannon introduced the concept of channel capacity, to measure the efficiency with which information is transmitted, and to find its upper limit.

      The capacity C of a channel: (information bit/symbol) is the maximum value of the mutual information I(X, Y) over the set of input symbols probabilities images

      [2.54] images

      The maximization of I(X, Y) is performed under the constraints that:

images

      The maximum value of I(X, Y)occurs for some well-defined values of these probabilities, which thus define a certain so-called secondary source.

      The capacity of the channel can also be related to the unit of time (bitrate Ct of the channel), in this case, one has:

      [2.55] images

      The channel redundancy Rc and the relative channel redundancy pc are defined by:

      [2.56] images

      [2.57] images

      The efficiency of the use of the channel images is defined by

      [2.58] images

      2.7.1. Shannon’s theorem: capacity of a communication system

      Shannon also formulated the capacity of a communication system by the following relation:

      [2.59] images

      where:

       – B: is the channel bandwidth, in hertz;

       – Ps: is the signal power, in watts;

        is the power spectral density of the (supposed) Gaussian and white noise in its frequency band B;

        is the noise power, in watts.

      EXAMPLE.– Binary symmetric channel (BSC).

      Any binary channel will be characterized by the noise matrix:

images

      If the binary channel is symmetric, then one has:

      p(y1/x2) = p(y2/x1) = p

      p(y1/x1) = p(y2/x2) = 1 − p

Schematic illustration of binary symmetric channel.

      Figure 2.5. Binary symmetric channel

      The channel capacity is:

images images

      Hence:

images

      But max H(Y) = 1 for p(y1) = p(y2). It follows from the symmetry of the channel that if p(y1) = p(y2), then p(x1) = p(x2) = 1/2, and C will be given by:

images Graph depicts variation of the capacity of a BSC according to p.

      Figure 2.6. Variation of the capacity of a BSC according to p

      The joined entropy of k random variables is written:

      [2.60] images

images

      One has:

      [2.61] images

      Equality occurs when the variables are independent.

      Конец ознакомительного фрагмента.

      Текст предоставлен ООО «ЛитРес».

      Прочитайте эту книгу целиком, купив полную легальную версию на ЛитРес.

      Безопасно оплатить книгу можно банковской картой Visa, MasterCard, Maestro, со счета мобильного телефона, с платежного терминала, в салоне МТС или Связной, через PayPal, WebMoney, Яндекс.Деньги, QIWI Кошелек, бонусными картами или другим удобным Вам способом.

/9j/4AAQSkZJRgABAQEBLAEsAAD/7SieUGhvdG9zaG9wIDMuMAA4QklNBAQAAAAAADAcAVoAAxsl RxwCAAACAAAcAlAAD1JhcGhhZWwgTUVOQVNDRRwCBQAITGF5b3V0IDE4QklNBCUAAAAAABDuBBKy CMqpqmejVZp8+vehOEJJTQQ6AAAAAADlAAAAEAAAAAEAAAAAAAtwcmludE91dHB1dAAAAAUAAAAA UHN0U2Jvb2wBAAAAAEludGVlbnVtAAAAAEludGUAAAAAQ2xybQAAAA9wcmludFNpeHRlZW5CaXRi b29sAAAAAAtwcmludGVyTmFtZVRFWFQAAAABAAAAAAAPcHJpbnRQcm9vZlNldHVwT2JqYwAAAAwA UAByAG8AbwBmACAAUwBlAHQAdQBwAAAAAAAKcHJvb2ZTZXR1cAAAAAEAAAAAQmx0bmVudW0AAAAM YnVpbHRpblByb29mAAAACXByb29mQ01ZSwA4QklNBDsAAAAAAi0AAAAQAAAAAQAAAAAAEnByaW50 T3V0cHV0T3B0aW9ucwAAABcAAAAAQ3B0bmJvb2wAAAAAAENsYnJib29sAAAAAABSZ3NNYm9vbAAA AAAAQ3JuQ2Jvb2wAAAAAAENudENib29sAAAAAABMYmxzYm9vbAAAAAAATmd0dmJvb2wAAAAAAEVt bERib29sAAAAAABJbnRyYm9vbAAAAAAAQmNrZ09iamMAAAABAAAAAAAAUkdCQwAAAAMAAAAAUmQg IGRvdWJAb+AAAAAAAAAAAABHcm4gZG91YkBv4AAAAAAAAAAAAEJsICBkb3ViQG/gAAAAAAAAAAAA QnJkVFVudEYjU

Скачать книгу