Dick Tracy's watch, back in the 1950s, allowed the comic strip detective to make phone calls from his wrist. Sure, he had a limited number of people he could call, but anyone watching him do it was impressed. Now, of course, cell phone technology has become ubiquitous, so the younger generations don't think much of Detective Tracy's cool watch. Today's equivalent might be a device allowing you to watch video streamed wirelessly to a palm-sized device from anywhere on the planet. This capability, as it emerges, will make hand held real-time video conferencing a reality.
So here we are 50 years after Dick Tracy and billions of telecommunications dollars later and we've solved many of the challenges of audio -- but what about wireless video? Most of the technology pieces are in place -- the only major issues remaining are consumer demand and an economic model that will make it feasible for companies to develop the solution. One of the most important technology pieces is the MPEG-4 video standard, and its early implementations. MPEG-4 is ideally suited to the task of wireless mobile device video, as will be discussed below. First, a bit of technology background that has led us to this point.
The technology for streaming audio by itself is acceptable - just consider music and phone conversations. Video, however, is different and, by the way, requires audio. In the 1920s silent movies gave way to movies with sound, but only after the technology advanced enough to make sound viable. Today, silent video is considered one shade shy of "useless" in the mind of the consumer. The two key challenges with adding sound to movies were "synchronization" and "amplification". In a streaming video environment either wireless or over a wired network, the basic problem of audio/video synchronization is still with us. We can tolerate some occasional glitches in video, but we are extremely sensitive to audio discrepancies such as stuttering or "out of sync" audio. The human ear can detect audio errors as small as a few milliseconds, so accurate audio and video synchronization is critical to successful video transmission.
To ship audio and video together either over a wired or wireless network, we must first employ techniques to put the video and audio in a container and keep it together before, during and after shipping. While much has been spoken and written concerning the techniques of encoding and decoding audio and video, the technologies involved in placing these encoded audio and video streams in containers for shipping over wired and wireless networks is less understood. It is our goal in this article to shed a bit of light on this technology aspect of audio and video transmission.
A/V Containers
So, what is an A/V container? Simply put, an A/V container is an information framework where encoded audio and video samples reside (see Figure 1). Another term to describe A/V containers is a "file format". A container is NOT an encoding/compression method. Many A/V containers are actually encoding/compression method agnostic. They simply describe how A/V samples coexist inside of a "file". AVI, MOV, ASF, MPEG-1 System Streams, MPEG-2 Program Streams, MPEG-2 Transport Streams, and MPEG-4 System Streams are all A/V containers.
Figure 1: A/V Container
The basic concept of an A/V container is that it is simply a "box" into which audio and video streams are placed for shipping to a destination. "Destinations" include residing on a hard drive as a file, a CD-ROM, a DVD, a remote location over a LAN or via the Internet. In some cases, the receiver has no idea what is in the container until it is opened. All container formats have "headers" that inform the receiver as to the contents of the container.
Figure 2: Audio/Video Encoder
Next: AVI Files and MPEG System Information
|
